OK, I skimmed the study. Just doing that, I can already point to problems. You ask about the source of their data -- well, it's a questionaire they mail around themselves.
Now, you linked a source about distribution of abortion clinics by ethnic neighborhood. That cites another paper with a description of the actual survey they did. Note this interesting difference: in the methodology / statistics paper ( http://onlinelibrary.wiley.com/doi/10.1363/4304111/pdf ), they say the following:
"The 378 specialized abortion clinics accounted
for 21% of all abortion providers, but performed 70% of
all abortions in 2008 (Table 4). Most of these facilities
reported 1,000 or more abortions during the year."
Looking at table 4, it's clear that "clinic" is a special category, not defined based on numbers. (In particular, while there are 378 clinics, there are 622 providers providing more than 400 abortions annually. Keep this number in mind).
Let's return now to your paper. They say, "Abortion clinics are defined as providers of 400 or more abortions annually."
In other words, they changed the definition of clinic between the source paper they cite _in the same paragraph_ and the geographic analysis paper they're doing at this point. Obvious terrible methodology that is extremely suspicious, and it took me 5 minutes to find that.
Oh, and you characterized this as "census data." Ha! This is THEIR census, not the US census.
Finally, you're the one who first brought up source quality, I believe, so it's perfectly fair to complain when you bring up a source of such terrible quality. Just because I may not have a source providing any number on this statistic does not mean I have to accept yours and its patently lopsided, terrible methodology.