Dear Arsenic Crisis Group

A preliminary assessment of the distribution of elevated arsenic (As)
in wells in Cambodia has shown a similar pattern to that observed
and reported by Dr Sengupta and his partners in W Bengal.

The elevated As appears to be closely associated with the levees of
the modern rivers, which I expect to consist of relatively coarse
grained sediments. The levees appear to represent the highest ground
in many areas and villages are often preferentially located on them.
The assoication is not 100%. While the highest levels of
contamination are found along the rivers, some areas with >50ug/L As
are found away from the river and developed in older alluvium.

It is clear that in Cambodia it should be possible to target high
risk areas for testing and mitigation, although I would suggest that
broader testing will also be required but perhaps less urgently. A
model of As distribution also allows us to target "High Risk" areas
in other parts of the country where little or no testing has been
undertaken to date.

What puzzles me however is why is the high As found in these
locations? My thoughts so far are:

I can rationalise the occurrence of As in the modern levee and
channel to some extent. The modern levee, and presumably the chanel
deposits of W Bengal are generally coarser grained and presumably low
in organic carbon. However, sandy deposits often have a relatively
low pore volume (say 30%) and in fact very little organic matter may
be required consume the dissolved oxygen and then to mobilise the
iron (+arsenic) - oxygen is not very soluble in water. Furthermore,
because these deposits are relatively permeable they are also
suitable for groundwater development, and are probably the "best"
aquifers based on flow.

But why not elevated As out on the floodplain? HIgh organic
contents, low groundwater permeability adds up to a high likelihood of
anoxic conditions suitable for As mobilisation. In addition little
flushing of released arsenic would occur because of relatively low
flows. The organic rich and clay rich levels of the flood plain
would not make good aquifers and so I would have thought that
some wells on the flood plain would have intersected old/buried
channels and levees.

Why don't these have high arsenic?

This is partly of only academic interest but partly it would help to target
future well testing and perhaps well drilling practices. Partly I
can't trust a model I don't understand.

David Fredericks