Scott Huddleston:
I googled "satellite maps of August 2010 Pakistan floods". Many relevant hits, here's one:
Thank you for this. That map gives me a clearer understanding of how the numerical mismeasure (of how much of Pakistan is "under water") might have come about. A good analogy here would be American weather maps which give warnings by county: Even though a tornado warning (say) may affect only a tiny corner of a given county, the entire county gets flagged. (Here in Canada, where northern counties can be quite large and irregularly shaped, this occasionally makes for some inappropriate/misleading pictures of where active weather may be occurring.) The "affected districts" map of Pakistan is like that: Even though only a tiny portion of the district may actually be under water, the area of the entire district gets added to the tally. Inattention to detail in news reporting takes care of the rest.