aInstituto de Investigaciones Oceanológicas, Universidad Autónoma de Baja California, Ensenada, Baja California,. México. bDepartamento de Ecologıa ...
Estuarine, Coastal and Shelf Science (1998) 46, 475–481
Cadmium in the Coastal Upwelling Area Adjacent to the California–Mexico Border J. A. Segovia-Zavalaa, F. Delgadillo-Hinojosaa and S. Alvarez-Borregob a
Instituto de Investigaciones Oceanolo´gicas, Universidad Auto´noma de Baja California, Ensenada, Baja California, Me´xico b Departamento de Ecologı´a, Centro de Investigacio´n Cientı´fica y de Educacio´n Superior de Ensenada, Ensenada, Baja California, Me´xico Received 20 March 1997 and accepted in revised form 26 September 1997 Cadmium concentrations ([Cd]) were measured in samples from the water column of the coastal upwelling zone adjacent to the California–Mexico border. Temperature and nutrient distributions showed an intense upwelling event during our sampling. Lowest [Cd] were found at locations offshore (50 km) (0·03–0·058 nM), whereas the maximum concentrations were found inshore (0·14–0·166 nM). Both nutrients and [Cd] were enriched in coastal waters. Our inshore [Cd] values are about 25% of those reported for waters off central California. This is possibly due to the intrusion of oligotrophic waters from the eastern edge of the North Pacific Central Gyre to the Southern California Bight. Multivariate analysis indicates that high [Cd]s were associated with high phytoplankton biomass, nutrients and low temperature. Our data present no evidence of a [Cd] gradient due to the San Diego and Tijuana sewage discharges, which indicates that they maintain a very local effect. 1998 Academic Press Limited Keywords: cadmium concentration; seawater; coastal upwelling; California–Mexico border
Introduction Coastal waters adjacent to the California–Mexico border receive large inputs of contaminants (mainly from the Point Loma outfall in San Diego, and from the City of Tijuana discharge). Heavy metals, such as cadmium (Cd), have high concentrations in these sewage (SCCWRP, 1990). However, San˜udoWilhelmy and Flegal (1991, 1996) estimated that 99% of Cd present in surface waters (1 m depth) of the California–Mexico border zone is related to physical processes, such as upwelling and advection, and only 1% to anthropogenic sources. A strong association of the vertical distribution of cadmium concentration ([Cd]) with that of inorganic nutrients, mainly phosphate and nitrate, has been recognized for two decades, suggesting that the oceanic biogeochemistry of this metal is controlled by the organic matter cycling processes (Bruland et al., 1994; de Baar et al., 1994 and others cited therein). Samples collected from the surfzone during 1991–93, near San Francisco Bay, confirm that upwelling largely determines temporal variations in [Cd] and nutrient concentrations (van Geen & Husby, 1996). Lares (1988) reported that, for locations at the southernmost limit of the Southern California Bight (SCB), Cd in the mussel Mytilus californianus has a 0272–7714/98/040475+07 $25.00/0/ec970296
seasonal variation with maximum concentrations during the upwelling season, similar to other locations of the California coast (Goldberg et al., 1983; and others cited therein). Lares and Orians (1997) also reported a similar pattern of [Cd] in mussels of the Canadian west coast. The objective of this work is to characterize and explain the distribution of [Cd] at different depths in the water column of the coastal upwelling zone adjacent to the California–Mexico border. We explored the covariation of [Cd], nutrients and phytoplankton biomass. Study area The California Current System (CCS) has a strong onshore component just south of the international border (Reid, 1988). In this area, the flow of the current divides; one component flows northward as the nearshore limb of the southern California Eddy, and the other flows southward along the Mexican coast. South of the border, inshore surface flow is equatorward most of the year (Lynn & Simpson, 1987). Pela´ez and McGowan (1986) analysed satellite colour imagery and found a latitudinally-oriented, sharp boundary just south of San Diego. This 1998 Academic Press Limited
476 J. A. Segovia-Zavala et al.
boundary coincides with the CCS onshore component and starts about 160 km off the coast, extending 500 km offshore. It is an abrupt transition between two large and different biological water masses with a three-fold change in phytoplankton pigment content over a distance of a few kilometres, with very low surface pigment concentration in the southern water mass. From June through winter the oligotrophic water mass makes an intrusion into the SCB. This low pigment area is located immediately offshore of the narrow coastal band of high pigment content, and it is continuous with the low pigment region south and west of the latitudinal boundary (Pela´ez & McGowan, 1986). Upwelling intensification and relaxation events occur during spring and summer on the Baja California coast, with very intense upwelling lasting about 4 days (Alvarez-Borrego & Alvarez-Borrego, 1982). These events are an important mechanism of metal transport (Bruland, 1980) and they produce strong surface [Cd] cross-shelf gradients (San˜udo-Wilhelmy & Flegal, 1991).
Materials and methods Sampling was done from the National University of Mexico’s RV El Puma, during 21–26 June 1990, in an area off north-western Baja California, adjacent to the California–Baja California border. A total of 35 hydrographic stations were occupied in lines perpendicular to the coast. Water samples for Cd analysis were taken at eight of the hydrographic stations (Figure 1). A Neil Brown CTD was used to generate salinity and temperature profiles. Reversible thermometers were also used on the sampling bottles. Water analyses were performed for salinity, dissolved oxygen (ml l 1), chlorophyll a (Chl a, mg m 3), nitrate (NO3, ìM) and phosphate (PO4, ìM). Chlorophyll a samples were immediately filtered through Gelman A/E (0·45 ìm) filters, extracted with 90% acetone for 24 h in the dark, and measured with a Turner 112 fluorometer following Holm-Hansen et al. (1965). Nutrients were analysed according to Parsons et al. (1984), with a Spectronic spectrophotometer. The Winkler method was used for oxygen following Parsons et al. (1984). Sampling, processing and storage of water samples for Cd analysis was done following the techniques described by Bruland et al. (1979) and Kremling et al. (1983). At each location, samples were taken from three depths. In all cases a sample was taken from 10 m, and depending on bottom depth, we sampled
40'
30'
20'
10'
50'
117°W San Diego
4 8
7 A6
5 3
40'
Point Loma
U.S.A.
2 1 Tijuana
30'
10 11 12 13 14 15
9
Coronados Is.
22
Mexico
21 20 19 18 17 16
20' Punta Descanso
B 23 0 5 10 Km
35
24 25 26 27 28 C
10'
34 33 32 31 30 29
Pta. Salsipuedes 32°N
F 1. Sampling locations. Hydrographic stations ( ); hydrographic stations where water was also sampled for Cd ( ). Broken lines A, B and C refer to transects shown in Figure 3.
from two more levels trying to cover the whole water column down to 150 m. Sampling was done from the bow, with GO-FLO bottles (General Oceanics). Water samples were taken from these bottles into precleaned 1 l polyethylene bottles. All shipboard sample handling was carried out in a Class-100 Environment laminar-flow clean-air bench. Cadmium was preconcentrated using an ammonium 1-pyrrolidine dithiocarbamate/diethylammonium diethyldithiocarbamate (APDC/DDDC) organic extraction (Bruland et al., 1979). Cadmium analysis was done with a Perkin–Elmer 5000 Atomic Absorption Spectrophotometer (Perkin–Elmer, Norwalk, CT) equipped with an HGA-500 Heated Graphite Atomizer. Analyses were performed in the Institute of Marine Science’s facilities of the University of California at Santa Cruz. The procedural blanks (x1 SD, n=5) were 0·00150·0004 nM. Results are equivalent to the dissolved fraction plus the particulate Cd leached into solution when samples were acidified for storage (Bruland et al., 1978). To explore the covariance structure of Cd and the other measured variables, we applied a Clustering Analysis (CA) and a Principal Components Analysis (PCA). Principal components were obtained from standardized variables because scales of the different properties have widely differing ranges (Popham & D’Auria, 1983). Cadmium concentration was not included as one of the variables in the PCA. Results of the PCA were input to a multiple regression analysis, with [Cd]s as the dependent variables and the principal components as independent variables.
Cadmium in the area off the California–Mexico border 477 20'
10'
Results and discussion
50'
17°W
Temperature at 10 m presented a very clear horizontal gradient, with values at the coast more than 5 C lower than offshore (Figure 2). We detected two plumes of cold water (13 C), one off the California–Mexico border, and the other off Punta Descanso. This horizontal T C gradient is clear evidence of upwelling and water-transport offshore. The vertical distributions of T C and PO4 show isograms raising onshore (Figure 3), and this was more intense in the southern part of our study area than in the northern part (Figure 3). The 13 C isotherm raised from 60 m at 50 km offshore to the surface inshore, in the southern part, while it only rose from 40 to 10 m in the northern part. The 0·75 ìM PO4 isogram had a similar behaviour to that of the 13 C isotherm (Figure 3). Phosphate had values greater than 1·0 ìM in waters over the continental shelf. The vertical distributions of T C and PO4 show that waters were moving from at least 200 m to shallower depths due to upwelling. Lowest [Cd] were found at locations furthest away from the coast (0·03–0·058 nM), while the maximum
San Diego
Point Loma
40'
19 18 17 16 15 14 13
30'
Punta Descanso
20'
13 10' 18
17 16
15
14 32°N
F 2. Temperature distribution (C) at 10 m.
A 1
0
2
3
6
7
B 8
9
19 17 15 13
50 Depth (m)
8
C 23 23
22
13
100
9
20
30
150
1
0
2
3
6
8
8
40
50 0
A 7
8
B 8
0.25 0.50
5
10
9
15
20
25
30
35 0
10
C 23 23
22
0.25
0.25
(a)
20
24
100
0
1.75
10 20 30 40 Distance offshore (Km)
28
1.0
1.25
0 50
1.50
1.25
100
1.25 1.50
1.50 2.0
26 27 0.75
1.25
PO4 µM
200 50 South
1.0
1.0
150
40
0.75
0.75 0.75
30
0.50
0.50
50 Depth (m)
10
10 10
8
200
50
9
10 0 North
0
15
11
9
150 200
13
28
11
10
T°C
25 26 27 17
15
11
100
24 19
19
17
150
1.50
1.75
(b)
1.75
50 0
5 10 15 20 25 30 Distance to the border (Km)
35 0
10 20 30 40 Distance offshore (Km)
200 50
F 3. Vertical distribution of temperature (upper panel), and phosphate (lower panel). The location of transects A, B and C are shown in Figure 1. Transect B is 50 km offshore.
478 J. A. Segovia-Zavala et al.
concentrations were found off the California–Mexico border (0·166 nM) and off Punta Descanso (0·14 nM), inshore (Figure 4). These [Cd] values are similar to those reported by San˜udo-Wilhelmy and Flegal (1996) for the Baja California coastal waters (5–45 km offshore) and for June 1988. The latter authors also sampled for [Cd] in the surfzone and found maximum values about 50% higher than our maximum. Maximum surfzone [Cd] values reported by van Geen and Husby (1996) for two sites near San Francisco Bay were more than four times our maximum (up to 0·8 nM). Comparing their values with vertical [Cd] profiles 100–200 km off San Francisco Bay (Bruland et al., 1978; Bruland, 1980), van Geen and Husby (1996) concluded that upwelled water came from as deep as 300 m. Our [Cd] values and those of San˜udo-Wilhelmy and Flegal (1996) are lower than van Geen and Husby’s (1996) because of the oligotrophic intrusion into the SCB. As mentioned above, this is an extension of the water with low surface photosynthetic pigment that comes from 500 km offshore to the coastal area (Pela´ez & McGowan, 1986). Bruland et al. (1994) reported [Cd] values of about 0·002 nM in the surface 100 m, and in the range 0·14–0·17 nM for 220–250 m depth, in the central North Pacific. Thus, the origin of upwelled water in our study area during the sampling period was 220–250 m depth at the eastern edge of the North Pacific Central Gyre. This implies that upwelling during April and May, before the oligotrophic intrusion into the SCB occurs, should produce [Cd] values in coastal waters of our study area closer to those reported for waters off central California. Previous studies have presented evidence of enrichment of coastal waters with trace elements in other regions of the world (Boyle et al., 1981; Bruland & Franks, 1983). Kremling (1985) reported a significant increase of [Cd], [Al], [Cu], [Mn] and [Ni] at the north-west European shelf edge area. A comparison with waters of the central North Sea indicated enrichment factors of about three to five for these metals. In our case there was a [Cd] enrichment factor of up to 5·5 for coastal waters with respect to offshore. Also, PO4 in coastal surface waters had an enrichment factor of up to three compared to concentrations in surface waters 50 km offshore (Figure 3). Thus, our results confirm that in our study area a process that contributes strongly to the enrichment of Cd in waters over the shelf is coastal upwelling. Since Figure 3 shows a north–south gradient of seawater properties over the shelf, it is interesting to analyse the behaviour of these properties in a transect parallel to the coast, 15 km offshore (Figure 5). The
vertical distribution of T C clearly shows the presence of a front [Figure 5(b)], located at about 25 km from the California–Mexico border, with the more stratified waters to the north. The distribution of non-conservative properties such as nutrients and Chl a are very much in accordance with the presence of this front. Nitrate, PO4 and Chl a from the surface to 20 m were much higher at 35–45 km from the border than in the northern part of the transect [Figure 5(c,e,f)]. In the southern extreme of the transect, Chl a was as high as 16 mg m 3, and NO3 was as high as 1·0 ìM, in surface waters, while they were only
