In Brazil we have our own legislation to arsenic and mercury, which are: arsenic = 1.0 mg / kg, 0.5 mg mercury / kg (for fish of most species and mercury 1.0 mg / kg for predatory fish. you can search larger data in the Codex Alimentarius.
Apart from the above references, many countries have their own standards for sediments or river water and even seawater also. continue your web search.
Ecological Soil Screening Levels (Eco-SSL) have been derived by the USEPA at the following sites: https://rais.ornl.gov/guidance/epa_eco.html and https://www.epa.gov/risk/ecological-soil-screening-level-eco-ssl-guidance-and-documents. Sediment Screening Values are available from the Massachusetts Department of Environmental Protection at http://www.mass.gov/eea/docs/dep/water/laws/a-thru-h/ecoturss.pdf. The latter document is for screening only, whereby if these values are exceeded then additional more comprehensive study is required.
I would say that there no standard for natural resources, only consumption has standard. Every source of water has particular standard that depend on the nature of passing entrance water on it.
Table 5.1: Heavy and trace metal concentrations in the CC as average of
Al
79780 mg/kga
Co
24 mg/kga
Sr
333 mg/kga
Ba
584 mg/kgh
Ti
4010 mg/kga
Ni
56 mg/kga
Mo
1.1 mg/kgc
Hg
40 µg/kgi
V
98 mg/kga
Cu
25 mg/kga
Ag
70 µg/kgd
Pb
14.8 mg/kga
Cr
126 mg/kga
Zn
65 mg/kga
Cd
100 µg/kge
Bi
85 µg/kge
Mn
716 mg/kga
As
1.7mg/kgb
Sn
2.3 mg/kgf
U
1.7 mg/kga
Fe
43975 mg/kga
Rb
78 mg/kga
Sb
300 µg/kgg
B
11 mg/kgj
a : UC: Shaw et al. (1967, 1976), LC: Rudnik and Presper (1990) in the proportions of standard profile of the continental crust derived from 3000 km long Standard Profile of Continental Crust (European Geo-Traverse) (EGT)) generalized on the basis of worldwide mapping, petrological studies and chemical balances.
b: UC calculated from rock average of Onishi and Sandell (1995), Burwash and Culbert (197 9) in relate to EGT profile mentioned before LC: gabbro, gneiss minus 20%..
c : UC: calculated from rock average compiled by Manheim and Landergeren (1978) in proportion to EGT profile, gabbro, gneiss minus 20% granite.
d: UC, LC: calculated from rock average of Hamaguchi and Kuroda, (1995) in proportion to EGT profile, Cu/Ag in felsic rocks 300 to 450.
e: UC, LC: calculated from rock average of Heinrichs et al. (1980) in proportion to EGT profile.
f: UC: calculated from rock average compiled by Hamaguchi and Kurdoda (1969) and Smith and Burton (1972) in proportion to EGT profile, LC: Rundick and Presper (1990).
g: UC: calculated from rock average of Onishi and Sandell (1995) and Burwash and Culbert (1979) in the proportion of EGT profile, LC: estimated.
h: UC: calculated from rock average complied by Puchelt (1972) in proportion to EGT profile, LC: Rudnik and Presper (1990).
i: UC, LC: calculated from rock average complied by the present author partly from Marowski and Wedepohl (1971).and partly from unpublished data in the proportion of EGT profile.
j: UC: calculated from rock average complied by Harder (1974) and Shaw et al. (1986) in proportion to EGT profile, LC: data from Truscott et al (1986), Leeman et al. (1992) and harder (1974).
I would also point out the sediment quality guidelines developed by NOAA, for use in estuarine and marine systems. (Long et al. 1998. Predicting toxicity in marine sediments with numerical sediment quality guidelines. Environmental Toxicology 17:714–727).
You can have a look to the recommended documents and use them as a guide, but you must remember that quality guidelines for sediments are dependent of the local geology. The suggestion from FS Kot to employ Turekian and Förstner and Whitman is a sound one.
Inigo's point (above) is a sound one, but the guidelines are usually based on biological impacts (toxicology) to benthic infauna. While local geology plays major role in trace element and metals concentration in sediments, if you care about biological endpoints/resource management, you can't just compare your data to the geology.
I agree with David. Inigo's post on geology would be helpful in seeking whether natural geological source are contributing to the heavy metal concentrations found at a site. Many regulations accept background concentrations of heavy metals, but require cleanup of man-made contamination if it exceeds human and ecological risk thresholds.
Your answer will depend on three things: 1) Do you have an analogous 'known clean' sample to compare to? 2) What are the specific chemical species of interest, and 3)What is the local jurisdiction and customary human impacts (say,from local agriculture) . If you have a baseline to compare to, the measured data variance from the baseline will give some idea of human impact.Beyond that, if the locality does not have guidelines, EPA or the equivalent E.U. agency guidelines can work as a standard.
The weight of trace metals are targeted according to Goncalves et al.,(1992) it is for Hg and Ni (1000 and 100) respectively and according to USEPA, (1976) and USPHS, (1962) are for As are (100), Ba (1.4), Cd (333), Cr (20), Cu (2), Pb (20), Ag (20), Zn (0.05), Co (20), Sr (0.2), Mo (14), Sn (20), Sb (200) and Bi (20).
These weights are calculated by taking the inverse proportion of the Maximum Permissible Level in surface water that used for domestic purposes.