Mytilus californianus as an indicator of heavymetals on the northwest coast of Baja

California, Mexico

A. MunÄ oz-Barbosa*, E.A. Gutie rrez-Galindo, G. Flores-MunÄ oz

Universidad AutoÂnoma de Baja California, Instituto de Investigaciones OceanoloÂgicas, PO Box 2516,

Ensenada, Baja California, Mexico

Received 8 August 1998; received in revised form 23 April 1999; accepted 25 May 1999

Abstract

Samples of Mytilus californianus were analyzed to determine the spatial and temporal dis-tribution of Ag, Al, Cd, Cr, Cu, Mn, Pb and Zn concentrations on the northwest coast ofBaja California and to compare the results with levels reported for California, USA. The

samples were collected at eight sites from the US±Mexican border to San QuintõÂ n, BC (300km south of the border) in February and July 1982 and 1994. During both years, north±southgradients in the concentration of Ag and Pb were observed, with the highest values close to

the US±Mexican border (Ag=0.60±1.54 and Pb=0.21±2.89 mg gÿ1) and the lowest in thesouthern part of the study area (Ag=0.05±0.10 and Pb=not detectable±0.03 mg gÿ1). The Agand Pb gradients suggest that their distributions are a�ected by anthropogenic activities closeto the US±Mexican border. South±north gradients in the concentration of Cd (mini-

mum=0.59±2.16 and maximum=2.61±12.93 mg gÿ1), for both 1982 and 1994, were observed.The geographic annual means of Pb for California (average of six stations along the entirestate) were signi®cantly higher than those of this study (average of the eight sampling sites) in

1982 and 1994. Cu in California was signi®cantly greater than that of this study only in 1994.Cluster analyses on all metals indicate three similar geographic zones in 1982 and 1994, sug-gesting a relative consistency in the processes that determine the spatial variability of the

concentration of some metals in M. californianus on the northwest coast of Baja California.# 2000 Elsevier Science Ltd. All rights reserved.

Keywords: Mytilus californianus; Heavy metals; Mexico

Marine Environmental Research 49 (2000) 123±144

www.elsevier.com/locate/marenvrev

0141-1136/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.

PI I : S0141-1136(99 )00052 -5

* Corresponding author.

1. Introduction

The coastal waters o� the California±Baja California border, on the southernlimit of the Southern California Bight, receive large amounts of pollutants fromurban areas in the USA (Los Angeles, San Diego) and Mexico (Tijuana, Ensenada)(SanÄ udo-Wilhelmy & Flegal, 1991). Wastewater from the treatment plants at PointLoma (San Diego) and Punta Bandera (Tijuana) particularly a�ect the border. ThePoint Loma treatment plant has an advanced system for the treatment of waste-water, and discharges it through a submarine pipe, 4 km from the coast at a depth of60 m. During 1992 this plant discharged an average of 704.1�106 l dayÿ1 and a totalof 17,870 t of suspended solids. From 1978 to 1992, it released an annual average of1.6, 1.1, 4.3, 15.8, 6.6, 10.4 and 30.4 t of Ag, Cd, Cr, Cu, Ni, Pb and Zn, respectively(SCCWRP, 1986, 1987, 1989a, 1992, 1994; Schafer, 1980, 1982, 1984). The PuntaBandera treatment plant has a secondary treatment system that discharges along thecoastline an average ¯ow of 130�106 l dayÿ1, and approximately 6000 t of sus-pended solids annually (SCCWRP, 1989b). This plant releases approximately 8.5,13.6, 1.8, 7.0, 6.6, 0.07 and 0.75 t yearÿ1 of Cu, Zn, Ni, Cr, Pb, Cd and Ag, respec-tively (Villaescusa-Celaya, Gutie rrez-Galindo & Flores-MunÄ oz, 1997). In order toevaluate the environmental impact of anthropogenic discharges on the spatial dis-tribution of heavy metals and identify other processes that determine their varia-bility in the coastal environment of Baja California, studies have been conducted onmetals in water (SanÄ udo-Wilhelmy & Flegal, 1991, 1992, 1994, 1996; Segovia-Zavala, Delgadillo-Hinojosa & Alvarez-Borrego, 1998) and in sediments (Gutie rrez-Galindo, Flores-MunÄ oz, Ortega-Lara & Villaescusa-Celaya, 1994; Villaescusa-Celaya et al., 1997). However, the study of metals in water and sediments provideslittle information on the bioavailability of these metals. The use of indicator organ-isms to identify areas polluted by heavy metals seems to be more attractive, since theorganisms not only concentrate the metals from the water, which allows for aneconomic and relatively easy analysis, but they also represent an average withrespect to time of the relative bioavailability of the metals in each study site (Phillips,1977). For more than two decades, mussels of the genus Mytilus have been proposedto evaluate the spatiotemporal trends of some pollutants in coastal environments(Beliae�, O'Connor, Daskalakis & Smith, 1997; Boalch, Chan & Taylor, 1981;Gault, Tollandand & Parker, 1983; Goldberg, 1975; Lauenstein, Robertson &O'Connor, 1990; Stephenson & Leonard, 1994). It is generally agreed that theseorganisms can accumulate the metals from their food and seawater to concentra-tions that considerably exceed those found in their environment; furthermore, theydo it in proportion to the concentrations found around them (Phillips, 1976a, b).Mussels have not been widely used as bioindicator organisms of heavy metals on

the northwest coast of Baja California and there has been no consistency withregard to the type and continuity of the studies made or the metals analyzed. Studieshave been made on the spatial variations of Pb (Chow, Snyder & Snyder, 1976),Ag (Martin, Stephenson, Smith, Gutierrez-Galindo & Flores-MunÄ oz, 1988), Hg(Gutie rrez-Galindo & Flores-MunÄ oz, 1986), and short-term temporal variations ofCd and Hg (Lares, 1988).

124 A. MunÄoz-Barbosa et al. /Marine Environmental Research 49 (2000) 123±144

The main objective of this work was to study the spatial and temporal variabilityin the concentration of Ag, Pb, Cd, Cu, Al, Mn, Zn and Cr inMytilus californianus, inthe early 1980s and mid-1990s, on the northwest coast of Baja California andidentify the principal processes involved in the distribution of these elements.Furthermore, the levels of metals found in this area are compared to those reportedin monitoring programs in California, USA.

2. Study area

The study area is the coastal zone from the US±Mexican border (San Diego±Tijuana) to San QuintõÂ n Bay, Baja California (300 km south of the border) (Fig. 1).The area lies within the southern part of the area known as the Southern CaliforniaBight, which extends, from east to west, from the coastline to the California CurrentSystem and, from north to south, from Point Conception, California, to Cabo Col-onet, Baja California (Mearns, 1973). Upwelling occurs throughout the year in thearea, with the most intense events from April to July. The hydrodynamics ofthe study area are complex. The current pattern is made up of a system in which theCalifornia Current dominates, transporting water towards the south (Lynn, Bliss &

Fig. 1. Study area and location of sampling sites.

A. MunÄoz-Barbosa et al. /Marine Environmental Research 49 (2000) 123±144 125

Eber, 1982). The principal inputs of fresh water to the coastal zone are the wastewaterdischarges from the cities of San Diego and Tijuana (Segovia-Zavala et al., 1998).

3. Materials and methods

3.1. Sampling

Specimens of M. californianus were collected in February and July 1982 and 1994from eight stations on the northwest coast of Baja California, Mexico (Fig. 1). Ateach sampling site, 45 and 90 organisms were collected during 1982 and 1994,respectively. Mussels of similar size (40±60 mm) were collected to minimize di�er-ences in the accumulation of trace metals due to age, size and variation in the ®l-tration rate (Boyden, 1977). Individuals were collected from a single mussel-bed andfrom the same tidal level to avoid variation in trace metal concentration due tophysiological di�erences caused by di�erent degrees of air exposure (Coleman,1980). The organisms were removed with a stainless steel knife (while the collector waswearing plastic gloves), placed in pre-cleaned plastic bags and frozen immediately.

3.2. Dissection

Samples were prepared for the analyses by thawing and cleaning the shells of epi-biotic debris. Forty-®ve mussels from each sample were divided into three groups of15 mussels each. The gonads were dissected and discarded to minimize variation inbody weight due to seasonality from the reproductive cycle. The length of the shelland weight of the somatic tissue were recorded individually. The tissue of the 15-organism sub-samples was homogenized at 45 rev minÿ1, in a Vitris 45 homogenizerequipped with titanium blades.

3.3. Digestion

An aliquot (�5.0 g wet wt.) of homogenized soft tissue was placed in 25-ml bea-kers with 5 ml of 70% trace metal grade HNO3, covered with a watch glass and leftovernight at room temperature. In the morning the sample was heated at 50�C forabout 2 h, afterwards it was re¯uxed for 4 h, taken to dryness and added with 1 mlof distilled deionized water (DDI). The sample was then slowly evaporated until theresidue was slightly moist, 20 ml of 1% HNO3 was added and the solution trans-ferred to an acid-cleaned 30-ml polyethylene vial. In order to determine the percent-age of humidity of the samples, approximately 1 g of each was put to dry at 70�Cfor 72 h.

3.4. Instrumental analysis

Pb, Ag and Cr analyses were conducted with a Termo Jarrel Ash Smith Hietfje 12atomic absorption spectrophotometer, equipped with CTF 188 graphite furnace.

126 A. MunÄoz-Barbosa et al. /Marine Environmental Research 49 (2000) 123±144

Air±acetylene ¯ame atomic absorption was used for Cd, Cu, Mn and Zn, and anitrous oxide-acetylene ¯ame was used for Al. Trace metals levels in NIST oystertissue (SRM-1566a) were measured to ensure the accuracy of analysis (Table 1).

3.5. Condition index

It has been observed that temporal ¯uctuations in the metal concentrations inmussels of the genus Mytilus are not necessarily related to the seasonal changesin the bioavailability of these metals, but rather to seasonal changes in the weight ofthe soft tissue of the organism (Marigo mez & Ireland, 1990); the latter could alsoa�ect spatial variations. Lares and Orians (1997) conducted normalizations bymultiplying the condition index (CI) by time series data of Pb and Cd concentrationsin M. californianus and found that this normalization reduces the e�ect of theexternal factors and nutritional state of the mussel. Using Cd and Pb data from apreliminary sampling conducted by Lares and Orians (1997), we performed corre-lations between normalizations with respect to the CI that involves organism volume(g dry wt./length�width�height) and length only (g dry wt./length). It was foundthat the correlations between both indexes are signi®cant (Pb � 0:95, p < 0:05;Cd � 0:89, p < 0:05). It can be concluded from the above that the normalizations:metal concentration�CI (g dry wt./length�width�height) and metal concen-tration�CI (g dry wt./length) are very similar and may be used indistinctly accord-ing to the criteria of the researcher and the available information. Only organismlength was available in this study, and the normalization used was that of metalconcentration�CI (g dry wt./length).

3.6. Statistics

Di�erences among sites and years of sampling were determined by a two-wayanalysis of variance (ANOVA) comparison on log-transformed data. Previous ana-lyses of normality (Kolmogorov±Smirnov) and homogeneity of variance (Barlett'stest) were performed on the log-transformed data as they are prerequisites of theseparametric analyses. All log-transformed data passed the tests of normality and

Table 1

Analysis of reference material oyster tissue SRM 1566a certi®ed for the National Institute of Standards

and Technology (mean�95% con®dence interval)

Measured Certi®ed

Ag 1.6�0.29 1.68�0.15

Al 157�10.9 202.5�12.5

Cd 3.90�0.18 4.15�0.38

Cr 1.58�0.50 1.43�0.46

Cu 58.8�4.01 66.3�4.3

Mn 11.15�0.47 12.3�1.5

Pb 0.352�0.015 0.371�0.014

Zn 771�19.4 830�57

A. MunÄoz-Barbosa et al. /Marine Environmental Research 49 (2000) 123±144 127

homogeneity of variance. Correlations among variables were computed from Pear-son's product±moment on the log-transformed data. In order to de®ne geographicalzones in the study area, hierarchical clustering was performed using dendrograms(complete linkage method) on the log10 of normalized metal concentrations (Metalconcentrations�CI).

3.7. Comparison of the data on trace metals in mussels from California with those ofBaja California

This consisted of comparing the geographic annual means (GAMs) of metal con-centrations in mussels from California in 1982 and 1995 (no data were available for1994) with the GAM of the study area of this work in 1982 and 1994. The GAMs ofthis study are the average of the eight stations, which, in turn, are the average of thevalues from February and July of a determined year. The GAMs of California wereobtained from data of the California State Mussel Watch (CSMW) in 1982 (Ladd etal., 1984) and by the National Status and Trends Program (NS&T) in 1995 andsome years between 1982 and 1995 (O'Connor & Beliae�, 1995). In order to obtainthe GAMs, six stations were used, de®ned as California Reference Stations (CRS)by CSMW in 1982. These stations are distributed from San Diego to Trinidad,California. The following stations were used from the CSMW program: TrinidadHead (010), Bodega Head (202), Paci®c Grove (414), Royal Palms (662), Oceanside(750) and Point Loma (840); and from NS&T: Eureka (HMBJ), Bodega Bay(BBBE), Paci®c Grove (PGLP), Palos Verdes (PVRP), Oceanside (OSBJ) and PointLoma (PLLH). Except at the Oceanside (OSBJ) station from NS&T, where thespecies Mytilus edulis was collected, resident mussels of the species M. californianuswere collected at all the stations of both programs.

4. Results

4.1. Ag

All heavy metal concentrations in M. californianus were calculated based on softtissue dry weight. In February and July 1982 and 1994, the Ag concentration in M.californianus presented a well-de®ned north±south gradient (Fig. 2). It was generallyobserved for the four samplings that the concentrations of Ag at the sites locatedmore to the north were signi®cantly greater ( p < 0:01) than those at the sites moretowards the south. In the four samplings, Ag presented signi®cant negative correla-tions ( p < 0:05) with Distance to Site 1 (DTS1) (r � ÿ0:877, ÿ0.835, ÿ0.969 andÿ0.895 for February and July 1982 and February and July 1994, respectively). Thisindicates that the Ag concentrations in M. californianus decreased as DTS1increased. This type of correlation has been reported in other works and is attrib-uted to the dilution and removal of Ag as the distance from its source increases(Martin et al., 1988; SanÄ udo-Wilhelmy & Flegal, 1992; Stephenson, Martin, Lange,Flegal & Martin, 1979). The concentrations of Ag in 1994 were signi®cantly

128 A. MunÄoz-Barbosa et al. /Marine Environmental Research 49 (2000) 123±144

( p < 0:01) greater than those of 1982. The highest concentration of the study wasobserved in July at Sampling Site 1 (SS1) (1.63 mg gÿ1) and the minimum in Feb-ruary at SS7 (0.01 mg gÿ1), both in 1994.

4.2. Pb

There was a north±south gradient in February 1982, but this was not so evident inJuly of the same year (Fig. 3). In February 1994, Pb concentration in M. california-

Fig. 2. Spatial variability of Ag concentrations in soft tissue of Mytilus californianus during (a) 1982 and

(b) 1994. Vertical bars represent �1SE and each point is the mean of three and six pooled samples (1982

and 1994, respectively) of 15 mussels each. *, February; *, July.

A. MunÄoz-Barbosa et al. /Marine Environmental Research 49 (2000) 123±144 129

nus at SS1 (0.54 mg gÿ1) was 4.5 times the sampling mean; with the exception of theformer value, this sampling generally recorded very low variability. There weresigni®cant positive correlations ( p < 0:05) between the Ag and Pb concentrations inM. californianus in three of the four samplings conducted (r � 0:904, 0.735, 0.901 forFebruary and July 1982 and February 1994, respectively). In July 1994, there wasconsiderable correlation (r � 0:510), but it was not signi®cant. SanÄ udo-Wilhelmyand Flegal (1992) found signi®cant correlations between Ag and Pb in seawater forthe same area. They state that this covariance is consistent with the similarities

Fig. 3. Spatial variability of Pb concentrations in soft tissue of Mytilus californianus during (a) 1982 and

(b) 1994. Vertical bars represent �1SE and each point is the mean of three and six pooled samples (1982

and 1994, respectively) of 15 mussels each. *, February; *, July. Note that vertical scales are di�erent.

130 A. MunÄoz-Barbosa et al. /Marine Environmental Research 49 (2000) 123±144

in the anthropogenic origin and a�nity of both metals to plankton. Thus, the spatialdistributions were regulated by similar anthropogenic inputs and removal processes.In general, Pb concentrations of the 1982 samplings are signi®cantly greater thanthose recorded for 1994 ( p < 0:01). The maximum concentration of the study wasobserved in February 1982 at SS3 (2.8 mg gÿ1) and the minimum in July 1994 at SS8(undetectable).

4.3. Cd

With the exception of SS7 and 8, whose Cd concentrations are low in February1982, the southern sites present signi®cantly higher concentrations ( p < 0:05) thanthose of the northern sites (Fig. 4). In general, Cd in M. californianus presented asouth±north gradient in February and July 1982 and 1994; however, this gradientwas not as noticeable in February 1982. The Cd distributions present a signi®cantinverse correlation (in some samplings) with those observed for Ag (r � ÿ0:752,ÿ0.877 and ÿ0.915 for July 1982 and February and July 1994, respectively) and Pb(r � ÿ0:780 for February 1994). This suggests that the controlling processes of Cdare not the same as those that control Ag and Pb. SanÄ udo-Wilhelmy and Flegal(1991) indicate that the di�erences between Pb and Cd in seawater are due mainly tothe fact that the processes that a�ect the distribution of Cd are di�erent from thoseof Pb. While the distribution of Pb in the ocean is controlled by atmosphericdeposition, the distribution of Cd is controlled by upwelling and advection. Themaximum concentration of the study was observed in July 1982 at SS6 (12.93 mggÿ1) and the minimum in February 1994 at SS1 (0.59 mg gÿ1).

4.4. Cu

No clear trend was observed in the geographic distribution of the Cu concentra-tion in M. californianus in February and July 1982 and 1994 (Fig. 5). The maximumconcentration of the study was recorded in February 1982 at SS3 (9.1 mg gÿ1) andthe minimum in July 1984 at SS6 (4.0 mg gÿ1).

4.5. Al, Mn, Zn and Cr

The concentrations of Al, Mn, Zn and Cr and the biometric variables, wet weight,size and CI did not present clear spatial or temporal trends, and for this reason theirdescription is omitted from the results. Tables 2 and 3 show the basic statistics of allthe metals studied in 1982 and 1994 (averages of February and July), respectively.In 1982, the dendrograms of the cluster analysis show three groups in both Feb-

ruary and July (Fig. 6a, b). In general, the stations are grouped according to therelative position in the study area. The northern stations are associated in onegroup, the central stations in another and the southern stations in the other. Some-thing very similar occurred in 1994. Three groups were also formed in February andin July (Fig. 6c, d) and the stations were associated according to their relative posi-tion in the study area, forming a northern group, a central group and a southern

A. MunÄoz-Barbosa et al. /Marine Environmental Research 49 (2000) 123±144 131

group. Thus, there are three geographic zones for the entire study area, ¯exible withregard to the sampling sites they include, but well de®ned.

4.6. Comparison of the data on trace metal concentration in mussels from Californiawith the concentration in Baja California

Even though the silver GAMs of California are greater than those of Baja Cali-fornia, 25.5 and nine times in 1982 and 1994, respectively (Table 4), based on the

Fig. 4. Spatial variability of Cd concentrations in soft tissue of Mytilus californianus during (a) 1982 and

(b) 1994. Vertical bars represent �1SE and each point is the mean of three and six pooled samples (1982

and 1994, respectively) of 15 mussels each. *, February; *, July.

132 A. MunÄoz-Barbosa et al. /Marine Environmental Research 49 (2000) 123±144

high variability found at the California stations in both years (SD � 11:45 and 8.22for 1982 and 1994, respectively) it was not possible to detect signi®cant di�erencesbetween the two zones. However, the di�erence between the means in 1982 indicatesthat at some of the California stations, the concentrations of Ag are greater thanthose found in Baja during the same year. In 1994 the di�erence between the meansis much lower, suggesting a decrease in Ag in California, even though at some sta-tions, such as Point Loma, the concentration (20.5 mg gÿ1 in 1995) continues to beconsiderably higher than that found at any station of Baja California. The GAMs of

Fig. 5. Spatial variability of Cu concentrations in soft tissue of Mytilus californianus during (a) 1982 and

(b) 1994. Vertical bars represent �1SE and each point is the mean of three and six pooled samples (1982

and 1994, respectively) of 15 mussels each. *, February; *, July.

A. MunÄoz-Barbosa et al. /Marine Environmental Research 49 (2000) 123±144 133

Table 2

February±July average metal concentrations (mg gÿ1 dry wt.) in the soft tissues of Mytilus californianus (mean�SD) in 1982 for each site of sampling a

Sites of sampling

1 2 3 4 5 6 7 8

Ag 0.716�0.203 0.432�0.275 0.225�0.166 0.136�0.050 0.068�0.019 0.088�0.018 0.110�0.079 0.089�0.021

(0.36±0.87) (0.19±0.95) (0.08±0.44) (0.10±0.22) (0.04±0.09) (0.06±0.11) (0.04±0.26) (0.05±0.122)

Pb 1.833�1.343 1.670�0.713 1.695�1.429 0.759�0.302 0.629�0.202 0.766�0.319 0.344�0.220 0.535�0.218

(0.74±4.35) (0.93±2.91) (0.33±3.80) (0.42±1.17) (0.39±1.00) (0.43±1.31) (0.05±0.56) (0.22±0.80)

Cd 2.03�0.79 2.30�0.32 3.56�0.90 5.58�1.28 5.71�0.50 10.11�3.53 7.20�4.58 8.23�4.28

(0.70±2.90) (1.80±2.60) (2.70±5.00) (4.40±7.70) (5.00±6.20) (7.10±16.00) (2.00±12.00) (3.80±13.80)

Cu 6.26�0.96 6.10�0.80 7.95�1.43 7.46�0.76 6.51�0.82 6.95�1.45 7.00�1.16 7.13�2.26

(5.3±7.5) (4.9±7.1) (6.5±9.7) (6.6±8.8) (5.4±7.9) (5.3±9.2) (5.5±8.2) (5.4±10.3)

Al 267�83 197�48 294�218 84�64 203�56 140�49 239�96 345�129

(160±379) (149±280) (107±632) (19±156) (114±281) (73±220) (149±367) (236±598)

Mn 11.46�4.20 12.33�2.90 9.38�5.13 9.18�2.97 10.32�2.28 10.12�3.01 12.85�4.49 11.75�4.24

(7.67±17.56) (8.07±15.40) (3.88±16.02) (6.09±12.05) (8.33±14.58) (6.65±15.15) (5.73±19.69) (7.81±20.08)

Zn 63�9 122�7 167�61 97�40 155�21 151�28 94�27 146�7

(51±76) (112±130) (103±258) (50±146) (125±184) (113±178) (54±131) (135±153)

Cr 1.618�0.334 1.912�0.267 1.764�0.486 1.128�0.129 1.576�0.396 0.732�0.077 1.484�0.354 1.549�0.212

(1.181±2.111) (1.530±2.317) (1.157±2.414) (0.878±1.307) (1.042±2.162) (0.653±0.841) (0.634±1.938) (1.146±1.863)

a Minimum and maximum concentrations are shown in parentheses.

134

A.MunÄ oz-B

arbo

saet

al./

Marin

eEnviro

nmentalResea

rch49(2000)123±144

Table 3

February±July average metal concentrations (mg gÿ1 dry wt.) in the soft tissues of Mytilus californianus (mean�SD) in 1994 for each site of sampling a

Sites of sampling

1 2 3 4 5 6 7 8

Ag 1.594�0.229 0.598�0.260 0.467�0.202 0.327�0.063 0.166�0.047 0.100�0.011 0.113�0.038 0.109�0.019

(1.194±1.952) (0.360±1.040) (0.227±0.790) (0.262±0.502) (0.078±0.227) (0.082±0.116) (0.054±0.178) (0.085±0.138)

Pb 0.333�0.216 0.107�0.037 0.038�0.025 0.150�0.080 0.025�0.023 0.031�0.031 0.081�0.048 0.015�0.024

(0.081±0.683) (0.039±0.177) (0.002±0.081) (0.038±0.272) (0.004±0.070) (0.002±0.070) (0.007±0.142) (ND±0.057)

Cd 0.61�0.13 1.20�0.51 1.55�0.49 3.16�2.05 3.05�0.76 6.06�1.35 6.36�4.02 2.50�1.21

(0.43±0.87) (0.67±2.26) (0.82±2.33) (0.91±6.54) (2.01±4.62) (4.21±7.70) (1.76±11.59) (1.43±5.04)

Cu 8.32�0.83 6.53�0.46 6.42�1.32 5.84�0.67 6.77�1.94 4.04�0.35 6.35�0.96 6.80�0.65

(6.77±9.18) (6.04±7.58) (4.82±9.66) (4.49±7.29) (4.18±10.34) (3.54±4.53) (5.01±7.99) (5.30±7.41)

Al 509�109 409�188 381�201 92�76 676�980 245�48 303�59 355�74

(307±743) (196±703) (163±616) (17±208) (328±992) (179±309) (227±412) (238±450)

Mn 7.94�0.64 5.86�1.60 5.45�1.36 4.07�1.31 19.84�5.56 6.94�1.26 9.50�2.57 7.09�0.64

(6.63±8.85) (3.95±8.06) (3.81±7.41) (2.68±6.07) (13.61±27.50) (5.57±8.44) (6.54±15.10) (5.98±7.98)

Zn 130�54 128�20 123�29 116�19 128�34 61�10 132�23 117�19

(67±244) (101±172) (74±176) (71±152) (75±171) (45±71) (91±183) (89±149)

Cr 0.674�0.459 1.067�0.297 1.804�1.342 0.558�0.168 1.309�0.247 1.074�0.246 1.091�0.370 1.497�0.397

(0.151±1.275) (0.788±1.578) (0.615±4.080) (0.393±0.795) (0.904±1.591) (0.803±1.472) (0.492±1.571) (0.951±2.053)

a Minimum and maximum concentrations are shown in parentheses.

A.MunÄ oz-B

arbo

saet

al./

Marin

eEnviro

nmentalResea

rch49(2000)123±144

135

Fig.6.Cluster

analysesaccordingto

log10ofmetals

concentrations�

conditionindex

(CI)

for(a)February

and(b)July

of1982and(c)February

and

(d)July

of1994.In

thisanalysis,Ag,Pb,Cd,Cu,Al,Mn,ZnandCrconcentrationswereused.

136 A. MunÄoz-Barbosa et al. /Marine Environmental Research 49 (2000) 123±144

Pb of California were signi®cantly higher than those of Baja California in 1982 and1994 (Table 4). The means of both zones decreased from 1982 to 1994, but the dif-ferences between them increased. The GAMs of California were 3.45 and 15.7 timesgreater than those of Baja California in 1982 and 1994, respectively. Although theconcentrations of Pb in the CRS are not high, some of them are much greater thanthose found in Baja California. No signi®cant di�erences were observed between theGAMs of Cd in California and Baja California in both years, and the same occurswith Cu in 1982. However, the GAM of Cu from California in 1994 is signi®cantlyhigher than that from Baja California.

4.7. Spatial distribution in California±Baja California

Fig. 7 shows the spatial distribution of Ag, Cd, Cu and Pb along California andBaja California. CRS data were used for California and the annual means of each ofthe sampling sites of this work for Baja California. The concentration of Ag in BajaCalifornia is close to the levels of the stations of northern California and is muchlower than that of the southernmost station (Point Loma) in 1982 and 1994 (Fig.7a). Point Loma is the station of the two zones that has the greatest concentration.From here two gradients are formed, one to the south (Baja California) and anotherto the north (central±northern California) with decreasing concentrations. On theother hand, the distribution of Cd (Fig. 7b) is completely di�erent from that of Ag:the low concentrations occur in northern Baja California (SS1±3) and southernCalifornia (Point Loma, Oceanside and Royal Palms) and the high concentrations inthe southernmost part of the study area of this work (SS4±8) and northern Cali-fornia (Paci®c Grove, Bodega Head, Trinidad Head) in both years. In this manner,the spatial distributions of both zones form a `U', where the high values at theextremities are similar to each other and the same occurs with the low values in

Table 4

Geographic annual means (GAMs), standard deviations, and results of t-test comparison for 1982

and 1994a

California (n � 6) This work (n � 8) t-test

GAM SD GAM SD

1982

Ag 5.88 11.45 0.23 0.22

Cd 4.88 4.18 5.59 2.86

Cu 6.48 2.59 6.92 0.61

Pb 3.52 2.69 1.02 0.59 0.02

1994

Ag 3.88 8.22 0.43 0.50

Cd 3.11 4.12 3.06 2.13

Cu 8.76 2.38 6.33 1.21 0.02

Pb 1.42 0.64 0.09 0.10 <0.01

a p values are shown when they are 40.05.

A. MunÄoz-Barbosa et al. /Marine Environmental Research 49 (2000) 123±144 137

the middle. In California, Pb has high concentrations at the stations of Paci®cGrove, Royal Palms and Point Loma in 1982 (Fig. 7c), and is signi®cantly di�erentfrom that found in Baja California during the same year (Table 4). In 1994, Pb inCalifornia decreased with respect to 1982, but it is signi®cantly di�erent from thatfound in this same year in Baja California (Table 4). Except for the concentrationsat Point Loma, which are the highest in 1982, 1994 and some years in between, thelevels of Cu in California and Baja California do not show clear spatial tendencies ineither of the two zones.

Fig. 7. California±Baja California spatial variability of (a) Ag, (b) Cd, (c) Pb, (d) Cu. *, 1982; &, 1994;

other symbols, years between 1982 and 1994.

138 A. MunÄoz-Barbosa et al. /Marine Environmental Research 49 (2000) 123±144

5. Discussion and conclusions

In studies conducted on sediments (Villaescusa-Celaya et al., 1997), water(SanÄ udo-Wilhelmy & Flegal, 1992) and organisms (Martin et al., 1988; Stephenson& Leonard, 1994), Ag has been associated with human activity. Ag is an excellenttracer of wastewater discharges; in general, the concentration decreases proportion-ally as the distance from the discharge increases. Martin et al. (1988) found an Agconcentration gradient in M. californianus from San Diego, California, to PuntaAbreojos, Baja California Sur. The high concentrations in the north were attributedmunicipal discharges. In a study conducted from the US±Mexican border to SanQuintõÂ n, Baja California, the Ag concentration in nearshore waters (<100 m)exhibited a north±south gradient and, according to a mass balance, the entire ¯owof Ag in the area can be derived from wastewater discharges from Point Loma, SanDiego (SanÄ udo-Wilhelmy & Flegal, 1992). These same authors also state that Pb:Agratios suggest that most of the Ag found in their study is principally from the waste-water discharges at Point Loma, San Diego. The most important inputs of Ag in theSouthern California Bight are municipal discharges (Schafer, 1976). In particular,the discharge of wastewater at Point Loma, San Diego, is the primary source of Agin these coastal waters and its in¯uence can be observed up to 250 km south of theborder (SanÄ udo-Wilhelmy and Flegal, 1992). The virtual absence of natural inputsof Ag in the area (SanÄ udo-Wilhelmy and Flegal, 1992) and the persistence of thehigh concentrations associated with areas close to densely populated zones (SS1)suggest that the distributions obtained in the four samplings conducted in this studycould be a product of anthropogenic inputs (Fig. 2). No signi®cant di�erences werefound between the GAM of Ag of California and that of Baja California in either ofthe two years (Table 4). However, in 1995 some places, such as Point Loma haveconcentrations 12 times greater than the highest concentration and 256 times greaterthan the lowest concentration detected in Baja California in 1994. This indicates thateven though the levels of Ag in mussels in some places of California have decreased,there are areas where the concentrations are still very high.As with Ag, the Pb concentration in seawater (Boyle, Sherrel & Bacon, 1994;

Bruland, 1983; Patterson, Settle & Glover, 1976), sediments (Villaescusa-Celaya etal., 1997) and organisms (Chow et al., 1976; Stephenson and Leonard, 1994) isdirectly linked to human activities. Pb enters the marine environment mainly fromthe atmosphere and Ag from wastewater. However, a common source and biogeo-chemical similarities could produce the covariation observed in this work, in spite oftheir di�erent form of access to the environment. Pb, as Ag, presented signi®cantnegative correlations ( p < 0:05) with DTS1. This type of correlation has beenreported in other works and it is attributed to the dilution and removal of Pb as thedistance from its source increases. In a study conducted in the same area, SanÄ udo-Wilhelmy and Flegal (1994) found a Pb gradient in seawater similar to that observedin the mussels in this study. They attributed this gradient partially to a net southern¯ow of the surface currents of the Southern California Bight during their samplings.They also indicate that, based on isotopic ratios, the Pb in seawater transportedtowards the south is a combination of local inputs of industrial Pb from the US and

A. MunÄoz-Barbosa et al. /Marine Environmental Research 49 (2000) 123±144 139

Mexico. The assimilation of Pb in mussels is a linear function of the Pb concentra-tion in the environment (Schulz-Baldes, 1974); therefore, the Pb concentration gra-dients in mussels could be a consequence of the Pb concentration in the seawater,which at the same time is a consequence of the combined anthropogenic inputs fromMexico and the US. Contrary to Ag, the GAM of Pb in California is signi®cantlygreater than that of Baja California in 1982 and 1994. The levels of Pb in the CRSdecreased from 1982 to 1985 (Table 4); however, they are still greater than those ofBaja California in both years. This is because the levels of Pb in Baja California alsodecreased. The decrease in Pb at some stations of California has been attributed tothe gradual elimination of leaded gasoline in the mid-1970s. In Baja California, thepartial implementation of unleaded gasoline began in 1990, and even though a long-term trend cannot be established from two points in time, the signi®cant decrease( p < 0:05) in Pb in the study area of this work from 1982 to 1994 could be attributedto it. Even though the decrease in Pb in mussels from California has been signi®cant,the concentrations found at some stations of California, such as Paci®c Grove in1995, are seven times greater than the highest value and 160 times greater than thelowest value found for Baja California in 1994.The biogeochemical cycle of Cd in the oceans is directly linked to organic matter.

The oceanic distribution of Cd is dominated by its involvement in a regenerationcycle similar to that of phosphates and nitrates; the maximum Cd concentrationscorrespond spatially with the maximums of these nutrients (Bruland, 1983). Alongthe coasts of California, USA, the Cd concentration in M. californianus was foundto present a spatial distribution that could not be attributed to anthropogenicinputs; the highest concentrations (10±13 mg ÿ1) were located at the San Miguel andSanta Barbara islands (Stephenson et al., 1979). These authors state that it may bebecause Cd has nutrient-like vertical pro®les and thus upwelling could carry highamounts of this element to the surface waters and this would result in the high levelsof metal in the organisms. In Baja California, high amounts of Cd are carried to thesurface waters by upwelling events. It has been found that these waters have up to®ve times more Cd than those 50 km o�shore (Segovia-Zavala et al., 1998). More-over, 99% of the Cd present in surface waters (1 m depth) on the northwest coast ofBaja California is related to physical processes, such as upwelling and advection,and 1% is related to anthropogenic sources (SanÄ udo-Wilhelmy & Flegal, 1991,1996). This suggests that the Cd concentration in mussels of the area is due toupwelling and not to anthropogenic sources. Nevertheless, it is not possibleto explain the spatial gradients of Cd found in the present study based on upwelling,since there are no data that could show a similar gradient in the intensity of thelatter. Besides, the gradients and temporal di�erences observed in this study mightlack validity, because of the short-term variability in the Cd concentration in M.californianus (Lares, 1988). However, the south±north gradients appear to be arelatively stable characteristic. This trend is observed in three of the four samplingsanalyzed in this study, and considering that there is a 12-year di�erence among someof the samplings, it can be concluded that there is consistency in the processes thatregulate Cd in the organisms. Something very similar also occurs in California,but in this case the gradient is north to south (Fig. 7b). As can be observed, this

140 A. MunÄoz-Barbosa et al. /Marine Environmental Research 49 (2000) 123±144

characteristic does not occur just in 1982 and 1995, but also in intermediate years.The spatial distribution of Cd in California±Baja California shows a trend that, inaccordance with the above, could also be relatively stable. The stations inside theSouthern California Bight (Royal Palms, Oceanside, Point Loma, SS1±5) showrelatively low levels of Cd, whereas the stations outside or at the limits of the bight(Trinidad Head, Bodega Head, Paci®c Grove, SS6±8) have high values. That is, forsome reason the mussels inside the bight have lower concentrations of Cd in theirtissue than those outside the bight.Some authors mention that mussels are not good bioindicators of Cu. Phillips

(1976a) showed that the net assimilation of Cu by M. edulis is a�ected by theoccurrence and change in the concentrations of Zn, Cd and Pb. Contrarily, the netassimilation of Zn, Cd or Pb does not seem to be a�ected by the presence of othermetals. The existence of this e�ect in M. edulis suggests that the organism is unsuit-able as an indicator of Cu contamination (Phillips, 1976b). Other studies indicatethat there is no obvious correlation between the Cu concentration in mussels andindustrial areas polluted with the same metal (Phillips, 1977). It appears that Cu ishandled atypically by the mussel, which may be a re¯ection of its extreme toxicity.In an experiment on mussel transplants, it was found that of all the metals exam-ined, Cu was one of those that most quickly decreased from high concentrationscaused by polluted waters to concentrations found in clean waters (Roesijadi et al.,1984). This may contribute to the inability of mussels as indicators of Cu and,therefore, cause Cu to not present a clear spatial trend in this study (Fig. 5), nor inthe California±Baja California distribution (Fig. 7). The GAM of Cu in Californiaseems to show a slight increase from 1982 to 1994, whereas that of Baja Califor-nia appears relatively stable in the same period (Table 4). This may be the cause forthe GAM of California to be signi®cantly greater than that of Baja California in1994. Notwithstanding, the di�erences between speci®c stations of California andBaja California are not as great as that of Ag and Pb. For example, Point Loma,which is the station that has the highest concentration in 1994, is 1.5 times greaterthan the highest value and three times greater than the lowest value detected in BajaCalifornia in 1994. Of the metals for which long-term temporal trends have beenobserved in California, Cu is the one that has occurred in many stations withincreasing trends (Stephenson & Leonard, 1994).The integration of three geographic zones in the cluster analysis is very probably

due to the metals that consistently occurred in spatial concentration gradients (Ag,Pb, Cd) in most of the samplings made. In the northern region of the study area,high concentrations were generally observed for Pb and Ag and low concentrationsof Cd. In the southern region, low concentrations were generally observed for Pband Ag and high ones for Cd. If we focus our attention on the distributions ofAg and Pb, these three zones could be considered as: an anthropogenically impactedzone (northern zone), moderately impacted (central zone) and another clean(southern zone).The spatial distributions of Ag and Pb found in this study suggest an anthropo-

genic in¯uence in the northern part of the study area. However, despite e�orts withregard to wastewater treatment in California, the levels found in Baja California are

A. MunÄoz-Barbosa et al. /Marine Environmental Research 49 (2000) 123±144 141

lower than those found at some places in California. On the other hand, the spatialdistribution of Cd does not show any anthropogenic in¯uence, but rather thespatial trends are completely opposite those of Pb and Ag. Cd is generally associatedwith upwelling; however, there are no data to support this a�rmation in this study.The spatial trend of Cd in the zone (south±north gradients) is apparently a stableand not an ephemeral characteristic, which is produced by natural phenomena andnot anthropogenic e�ects. The concentrations of Cd in Baja California and Califor-nia are very similar. However, the graphic comparison (Fig. 7) reveals what may alsobe a non-ephemeral characteristic: the stations inside the Southern California Bightshow lower concentrations than those outside or at the limits of the bight. Cu doesnot show clear spatial trends and the concentrations found, albeit signi®cantly lowerthan those of California in 1994 (Table 4), do not show di�erences with speci®c sitesof California as large as those of Pb and Ag.The relative consistency of the three geographic zones de®ned support the stability

of the spatial trends of Ag, Cd and Pb and suggests that the processes that determinethe spatial variability of the concentration of some metals are also relatively con-sistent. The fact that signi®cant decreases or increases are observed in the con-centration of some metals in M. californianus with respect to time, they are probablydue to a decrease or increase in the intensity of the same processes, also with respectto time. That is, quantitatively, there are variations in some cases, but qualitativelythe general picture remains relatively constant.

Acknowledgements

We are grateful to Dr S.A. Sa¯udo-Wilhelmy and two anonymous reviewers fortheir critical comments on the manuscript.

References

Beliae�, B., O'Connor, T. P., Daskalakis, D. K., & Smith, P. J. (1997). U.S. mussel watch data from 1986

to 1994: temporal trend detection at large spatial scales. Environ. Sci. Technol., 37, 1411±1415.

Boalch, R., Chan, S., & Taylor, D. (1981). Seasonal variation in the trace metal content of Mytilus edulis.

Mar. Pollut. Bull., 12, 276±280.

Boyden, C. R. (1977). E�ect of size upon metal content of shell®sh. J. Mar. Biol. Ass. UK, 57, 675±714.

Boyle, E. A., Sherrel, R. M., & Bacon, M. P. (1994). Lead variability in the western North Atlantic Ocean

and central Greenland ice: implications for the search for decadal trends in anthropogenic emissions.

Geochim. Cosmochim. Acta, 58, 3227±3238.

Bruland, K. W. (1983). Trace elements in sea-water. In J. P. Riley, & R. Chester, Chemical oceanography

(pp. 157±220). London: Academic Press.

Chow, T. J., Snyder, H. G., & Snyder, C. B. (1976). Mussels (Mytilus sp.) as an indicator of lead pollu-

tion. Sci. Total. Environ., 6, 55±63.

Coleman, N. (1980). The e�ect of immersion on cadmium accumulation by Mytilus edulis. Mar. Pollut.

Bull., 11, 359±362.

Gault, N. F. S., Tollandand, E. L. C., & Parker, J. G. (1983). Spatial and temporal trends in heavy metal

concentrations in mussels from Northern Ireland coastal waters. Mar. Biol., 77, 307±316.

142 A. MunÄoz-Barbosa et al. /Marine Environmental Research 49 (2000) 123±144

Goldberg, E. D. (1975). The mussel watchÐa ®rst step in the global marine monitoring. Mar. Pollut.

Bull., 6, 111.

Gutie rrez-Galindo, E. A., & Flores-MunÄ oz, G. (1986). Biological availability of mercury in coastal waters

of northern Baja California. Cienc. Mar., 12, 85±89.

Gutie rrez-Galindo, E. A., Flores-MunÄ oz, G., Ortega-Lara, V., & Villaescusa-Celaya, J. A. (1994). Heavy

metals in coastal sediments of the Baja California (Me xico)±California (USA) border zone. Cienc.

Mar., 20, 105±124.

Ladd, J. M., Hayes, S. P., Martin, M., Stephenson, M. D., Coale, S. L., Lin®eld, J., & Brown, M. (1984).

California State Mussel Watch: 1981±83 (Water Quality Monitoring Report No. 83-6TS). Sacramento,

CA: State Water Resources Control Board.

Lares, M. L. (1988). VariacioÂn temporal de cadmio y mercurio biodisponibles en una zona de surgencias

costera. MS thesis, CICESE, Ensenada, BC, Mexico.

Lares, M. L., & Orians, K. S. (1997). Natural Cd and Pb variations in Mytilus californianus during the

upwelling season. Sci. Total. Environ., 197, 177±195.

Lauenstein, G. G., Robertson, A., & O'Connor, T. P. (1990). Comparison of trace metal data in mussels

and oysters from a Mussel Watch Program of the 1970's with those from 1980's program. Mar. Pollut.

Bull., 21, 440±447.

Lynn, R. J., Bliss, A., & Eber, L. E. (1982). Vertical and horizontal distributions of seasonal mean tem-

perature, salinity, sigma-t, stability, dynamic height, oxygen and oxygen saturation in the California

Current, 1950±1978. CalCOFI Atlas 30.

Marigo mez, I., & Ireland, M. P. (1990). A laboratory study of cadmium exposure in Littorina littorea in

relation to environmental cadmium and exposure time. Sci. Total. Environ., 90, 75±87.

Martin, M., Stephenson, M. D., Smith, D. R., Gutierrez-Galindo, E. A., & Flores-MunÄ oz, G. (1988). Use

of silver in mussels as a tracer of domestic wastewater discharge. Mar. Pollut. Bull., 9, 512±520.

Mearns, A. J. (1973). Southern California inshore demersal ®shes: diversity, distribution and disease as

response to environmental quality. CalCOFI Report XVII.

O'Connor, T. P., & Beliae�, B. (1995). Recent trends in coastal environmental quality: results from the

Mussel Watch Project 1986 to 1993 (Appendix). Silver Spring, MD: NOAA N/ORCA, National Status

and Trends Program.

Patterson, C. C., Settle, D. M., & Glover, B. (1976). Analisys of lead in polluted coastal seawater. Mar.

Chem., 4, 305±319.

Phillips, D. J. H. (1976a). The commonmusselMytilus edulis as an indicator of pollution by zinc, cadmium,

lead and copper. I E�ects of environmental variables on uptake of metals. Mar. Biol., 38, 59±69.

Phillips, D. J. H. (1976b). The common mussel Mytilus edulis as an indicator of pollution by zinc, cad-

mium, lead and copper. II Relationship of metals in the mussel to those discharged by industry. Mar.

Biol., 38, 71±80.

Phillips, D. J. H. (1977). The use of biological indicator organisms to monitor trace metal pollution in

marine and estuarine environmentsÐa review. Environ. Pollut., 13, 281±317.

Roesijadi, G., Young, J. S., Drum, A. S., & Gurtisen, J. M. (1984). Behavior of trace metals in Mytilus

edulis during a reciprocal transplant ®eld experiment. Mar. Ecol. Prog. Ser., 18, 155±170.

SanÄ udo-Wilhelmy, S. A., & Flegal, A. R. (1991). Trace elements distributions in coastal waters along the

US±Mexican boundary: relative contributions of natural processes vs. anthropogenic inputs. Mar.

Chem., 33, 371±392.

SanÄ udo-Wilhelmy, S. A., & Flegal, A. R. (1992). Anthropogenic silver in the Southern California Bight: a

new tracer of sewage in coastal waters. Environ. Sci. Technol., 6, 2147±2151.

SanÄ udo-Wilhelmy, S. A., & Flegal, A. R. (1994). Temporal variations in lead concentrations and isotopic

compositions in the Southern California Bight. Geochim. Cosmochim. Acta, 58, 3315±3320.

SanÄ udo-Wilhelmy, S. A., & Flegal, A. R. (1996). Trace metal concentrations in the surf zone in coastal

waters o� Baja California. Environ. Sci. Technol., 30, 1575±1580.

SCCWRP (1986). Characteristics of municipal wastewater. Southern California Coastal Water Research

Project. Annual Report, 1986. Long Beach, CA.

SCCWRP (1987). Coastal Water Research Project: Southern California Coastal Water Research Project.

Annual Report, 1987. El Segundo, CA.

A. MunÄoz-Barbosa et al. /Marine Environmental Research 49 (2000) 123±144 143

SCCWRP (1989a). Recent changes and long term trends in the combined mass emission discharged into

the Southern California Bight. Southern California Coastal Water Research Project, Annual Report,

1988. Long Beach, CA.

SCCWRP (1989b). Marine outfall: 1987 inputs from wastewater treatment plants, power plants and

industrial facilities. Southern California Coastal Water Research Project. Annual Report, 1988. Long

Beach, CA.

SCCWRP (1992). Characteristics of e�uents from large municipal wastewater treatment facilities in 1990

and 1991. Southern California Coastal Water Research Project. Annual Report, 1990±91 and 1991±92.

Long Beach, CA.

SCCWRP (1994). Characteristics of e�uents from large municipal wastewater treatment facilities in 1992.

Southern California Coastal Water Research Project. Annual Report, 1992±93. Long Beach, CA.

Schafer, H. A. (1976). Characteristics of municipal wastewater discharges. Southern California Coastal

Water Research Project. Annual Report, 1976. El Segundo, CA.

Schafer, H. A. (1980). Characteristics of municipal wastewater. Southern California Coastal Water

Research Project. Biennial Report, 1979±1980. Long Beach, CA.

Schafer, H. A. (1982). Characteristics of municipal wastewater. Southern California Coastal Water

Research Project. Biennial Report, 1981±1982. Long Beach, CA.

Schafer, H. A. (1984). Characteristics of municipal wastewater. Southern California Coastal Water

Research Project. Biennial Report, 1983±1984. Long Beach, CA.

Schulz-Baldes, M. (1974). Lead uptake from seawater and food, and lead loss in the common mussel,

Mytilus edulis. Mar. Biol., 25, 177±193.

Segovia-Zavala, J. A., Delgadillo-Hinojosa, F., & Alvarez-Borrego, S. (1998). Cadmium in the coastal

upwelling area adjacent to the California±Mexico border. Estuar. Coastal. Shelf. Sci., 46, 475±481.

Stephenson, M. D., Martin, M., Lange, S. E., Flegal, A. R., & Martin, J. H. (1979). Trace metal con-

centration in the California mussel Mytilus californianus (California Mussel Watch 1977±1978, Water

Quality Monitoring Report, Vol. II).

Stephenson, M. D., & Leonard, G. H. (1994). Evidence for the decline of silver and lead and the increase

of copper from 1977 to 1990 in the coastal marine waters of California. Mar. Pollut. Bull., 28, 148±153.

Villaescusa-Celaya, J. A., Gutie rrez-Galindo, E. A., & Flores-MunÄ oz, G. (1997). Heavy metals in geo-

chemical sediments fractions of the border region between Baja California, Me xico, and California,

USA. Cienc. Mar., 23, 43±70.

144 A. MunÄoz-Barbosa et al. /Marine Environmental Research 49 (2000) 123±144