If you are interested in this subject, send me an email (firstname.lastname@example.org) or leave a comment.
While it is impossible to identify the exact location of subsurface drains in the conterminous U.S. (Sugg, Jayne, Naz, Brown), this open community is focusing on the inferring the presence of tile drains on agricultural lands. These pages provide a forum to collaborate about methods, tools, and data sets useful for at least the conterminous United States (CONUS). Based on literature reviews, the group members are not aware of previous endeavors to assemble such a dataset for CONUS with the most recent available data (SSURGO, CDL, NLCD) (apart from Wieczorek’s (USGS) work with the 1992 NRI data).
Although some data on the national scale exists in tabular form, such as Ag Census and NRI, many researchers struggle with the limitations of this content. Perceived limitations include methods of data collection, precision, and initial definitions of variables used in analysis (for example, what constitutes poorly drained soils) and the identification of the differences between types of drainage (surface versus tiles). Because of these limitations, defining tile drains using spatial analysis is even more difficult than probabilistic methods (but that do not give as precise location).
Nakagaki, N., and Wieczorek, M.E., 2016, Estimates of Subsurface Tile Drainage Extent for 12 Midwest States, 2012: U.S. Geological Survey data release, http://dx.doi.org/10.5066/F7W37TDP
Nakagaki, N., Wieczorek, M.E., and Qi, S.L., 2016, Estimates of Subsurface Tile Drainage Extent for the Conterminous United States, early 1990s: U.S. Geological Survey data release, http://dx.doi.org/10.5066/F7RB72QS\
Sugg, Zachary, Assessing U.S. Farm Drainage, Can GIS Lead to Better Estimates of Subsurface Drainage Extent? Sugg's methodology is an attempt to establish ranges of tile drainage extent at the state level.
Other information useful/needed to estimate historically and currently tiled areas?
Can we use historic maps and historical records (1969, 1974 and 2012) to tease out anything spatially?
Does knowing specific crop types help mask out areas? Would a more precise coverage of wetlands help?
What are the impacts on modeling efforts such as SPARROW or WARP when using newer estimates of Tile drains?
1992 NRI apportioned data was used in Cycle 2 what will be the "official" USGS Tile Drain estimate to be used byall components of NAWQA in Cycle 3? and how will it be verified as sufficient?
How to choose the correct vertical averaging technique? Should this vary as a function of drainage class?
Are there finer-resolution data that might aid in zeroing in on potential areas ripe for tiling?
For example: Soil restrictive layer, Drainage Class, Hydrologic group, minimum water table depth, hydric rating, Corestrictive Layer, et cetera. Some illustrations of different variables are shown below.
Mike Wieczorek (USGS)