...I forgot who said it, perhaps they were wrong? This stuff is well beyond my areas of knowledge.
t'was me.
the real answer, the same answer that any complicated question gets is: it depends...
i saw the same piece, and it was somewhat nuanced, though i fear in a way that the journalists didn't understand, and therefore got wrong - the 'baseline' is that Chemical weapon agents will be destroyed by intense heat (burning), however there may well be peices of ordinance or continers full of this stuff that don't get the 'full' treatment of the incoming weapons effects and so therefore don't get destroyed. the calculation/gamble is over what happens to the CW that don't get destroyed, but are in a place full of fire and whizzing bits of metal.
now, when you use a chemical weapon you don't just open a barrel of this stuff and let the wind do its job - you disperse it using munitions or systems not unlike a crops-praying aircraft, and you try to achieve a happy medium of dosage vs coverage.
personally my view is that if you were to have an attack on a CW facility, but where some of it was not destroyed but opened to the elements, the result would be a very heavy contamination (far beyond fatal) in a very small area (i'm hugely dubious about this '1-2km' idea..), and very light contamination over a much larger area where the undestroyed CW agents get thrown into the air by smoke/heat.
my basis for this is our own planning for CW attacks by Warsaw Pact forces, and our planning for responding to accidents involving CW.
certianly, if on our inevitable march east over the crumpled remains of the Red Army had it all gone hot in the late 1980's, we would not have deliberately driven through wrecked Soviet CW armed Artillery regiments, but we would not have avoided them by a mile either. that said, the 'heat destroys CW' thing also meant that along with airfeilds, bridgeheads, Logs bases, HQ's and Bridges, CW dumps were due for a visit from a bucket of instant sunshine...