Chromium, iron, nickel, copper, zinc and lead
were analysed using an external standard calibration
curve.
2.5. Inducti¨ely-coupled plasma emission
spectrometry
The digested samples were analysed in triplicate,
using an ARL 35000-C model with automatic
sampler, for analysis of cadmium in water
and sediment samples. Cadmium concentrations
in water were found below the ICP detection
limit -5 mgrl.. The cadmium concentration was
measured according to certified materials CRM
277 and Soil-7 Table 1..
2.6. Statistical methods
Data were processed by the SAX program, in
order to obtain average values, standard deviations
and confidence limits. The SASW software
SAS Institute Inc, 1989, 1990. allowed an investigation
of the existence of a normal distribution
for the average values. With this program, the
variance analysis ANOVA. is applied to the average
values. In order to deduce temporal and
spatial distributions for heavy metals in the Alzate
Reservoir, Duncan distribution, and Kruskal]
Wallis test are used when no normal distribution
rule was observed. Moreover, regression and
Pearson multiple correlation analysis are applied
to all the results. The different statistical methods
were performed with a 95% confidence as0.05..
3. Results and discussion
3.1. Water metal concentrations
The average level of metals in water are illustrated
in Figs. 2]5. Table 2 shows minimum,
maximum, average values and significant differences
of heavy metal concentrations for all sampling
sites, for all campaigns. In general the heavy
metal concentrations of the water were found to
decrease in the sequence: Fe)Hg)Cr)Cu)
Zn)Pb)Ni. Zinc, copper and nickel concentra