UC Davis

Using a critical cartography/GIS approach and multiple data sets on community supported agriculture (CSA), this paper addresses two questions. First, how accurate was the 2007 United States Census of Agriculture's counting of CSAs? Second, where are CSAs concentrated and how does their distribution compare with that of farming generally and population distribution? I argue that significant overcounting of CSAs in the census occurred largely because of a lack of shared meanings of terms. Examination of discrepancies between data sets at the county level in California points to higher CSA overcounting in counties with many farms and lower overcounting in counties with many CSAs. This overcounting matters as critical geographers and others increasingly seek to take stock of and contribute to alternative agrifood movements. As for distribution, multi-scale maps and a CSA density indicator reveal the continued existence of high levels of CSAs in New England and the Pacific Northwest and low levels in the South. Further data collection and analysis efforts should focus on (1) improving census questions for the next CSA count, including adding a definition of CSA and asking whether farmers coordinate the CSA or contribute to a joint CSA; and (2) conducting further geographical analyses on CSA vis-à-vis contributing causes behind low and high concentrations and areas of fast and slow growth.