Keys for merging US state/county ANSI's with GADM Unique identifiers?

Question

I've got county-level data from USDA NASS, and I want to merge it with gridded climate data. A preliminary step is the associate the counties with the coordinates of their centroids. If I merge by names, about 5% of the counties are left out, because of different representations of names. For example, the city and county of Chesapeake, Virginia is listed as "Chesapeake" in GADM and "Chesapeake City" in the NASS data.

The NASS data has ANSI codes, while the GADM data has its own scheme. Is there some sort of concordance table or dictionary that allows me to link these two?

Here is an example of the NASS data: http://quickstats.nass.usda.gov/results/E03CA5B5-AE87-3C9D-B8E7-662F85AD3CC0

And here is an example of the GADM data:

2> library(raster)
2> adm <- getData('GADM', country='USA', level=2)
trying URL 'http://biogeo.ucdavis.edu/data/gadm2.7/rds/USA_adm2.rds'
Content type 'unknown' length 13936198 bytes (13.3 MB)
==================================================
downloaded 13.3 MB

2> head(adm)
  OBJECTID ID_0 ISO        NAME_0 ID_1  NAME_1 ID_2  NAME_2 TYPE_2 ENGTYPE_2
1        1  244 USA United States    1 Alabama    1 Autauga County    County
2        2  244 USA United States    1 Alabama    2 Baldwin County    County
3        3  244 USA United States    1 Alabama    3 Barbour County    County
4        4  244 USA United States    1 Alabama    4    Bibb County    County
5        5  244 USA United States    1 Alabama    5  Blount County    County
6        6  244 USA United States    1 Alabama    6 Bullock County    County
  NL_NAME_2 VARNAME_2
1        NA      <NA>
2        NA      <NA>
3        NA      <NA>
4        NA      <NA>
5        NA      <NA>
6        NA      <NA>

The county ANSI's don't appear to correspond to the GADM's ID_2 field.