For sediments, normalization with source area values would be more meaningful if available (uncontaminated river sediments in case of coastal marine sediments). In a sedimentary record of unconsolidated sediments (a core for example), values of deeper sediments which would have not been influenced by anthropogenic activity would be useful.

In absence of background values/source area composition shale values are frequently employed.

Sometimes, Igeo and other indices are taken too far because of the choice of values used for normalization. For example, sediments from a given area depending on the source lithology may have 40ppm of Pb (geogenic), which is higher than shale (20ppm) or the upper crust (~13ppm) but not necessarily contaminated.

While normalisation to shale averages (e.g. North American Shale Composite, NASC; Post-Archean average Australian Shale, PAAS) is commonly done to assess trace-element enrichment or depletion for sedimentary rocks, it's probably best to use local background concentrations for soils.

The rationale for using shales is that they integrate crustal material derived from different earth-surface environments, so their deposition and formation is in itself an averaging process (an assumption supported by the consistency in their composition across different shales).

For sediments, normalization with source area values would be more meaningful if available (uncontaminated river sediments in case of coastal marine sediments). In a sedimentary record of unconsolidated sediments (a core for example), values of deeper sediments which would have not been influenced by anthropogenic activity would be useful.

In absence of background values/source area composition shale values are frequently employed.

Sometimes, Igeo and other indices are taken too far because of the choice of values used for normalization. For example, sediments from a given area depending on the source lithology may have 40ppm of Pb (geogenic), which is higher than shale (20ppm) or the upper crust (~13ppm) but not necessarily contaminated.

I also favour the answers given by prof.Andrew and Nagender. It is meaningful for us when we take the local values (unpolluted region values) as baseline values.

Dear Srujana,

The relevance of any index or normalizer depend on what one hopes to achieve through its use. If you spell out exactly the problem you are trying to solve using a geo-accumulation index, you might get more feedback.

Turekian KK, Wedepohl KH. Distribution of the elements in some major units of the earth’s crust. Bull Geol Soc America. 1961;72:175–92