Course Proximity of Water to NLCD Agriculture

The National Hydrography Dataset forms the basis of the water body margin dataset.

The National Land Cover Dataset was compiled from Landsat satellite TM imagery (circa 1992) with a spatial resolution of 30 meters and supplemented by various ancillary data (where available). The analysis and interpretation of the satellite imagery was conducted using very large, sometimes multi-state image mosaics (i.e. up to 18 Landsat scenes). Using a relatively small number of aerial photographs for 'ground truth', the thematic interpretations were necessarily conducted from a spatially-broad perspective. Furthermore, the accuracy assessments (see below) correspond to 'federal regions' which are groupings of contiguous States. Thus, the reliability of the data is greatest at the State or multi-State level. The statistical accuracy of the data is known only for the region. 
The following NLCD categories compose the agriculture class: 82. Row Crops - Areas used for the production of crops, such as corn, soybeans, vegetables, tobacco, and cotton. 83. Small Grains - Areas used for the production of graminoid crops such as wheat, barley, oats, and rice 84. Fallow - Areas used for the production of crops that are temporarily barren or with sparse vegetative cover as a result of being tilled in a management practice that incorporates prescribed alternation between cropping and tillage.

Watershed boundaries upon which data is summarized

Lineage/Process Step: 
(1) Create subregion NHD geodatabases 

USGS NHD Append 2.25 AMLs were used to aggregate Huc-08 NHDinArc hydrology to the Huc-04 subregion level. 

(2) Project NHD 

The resulting coverages were then projected from decimal degrees to Albers Conical Equal Area using ArcInfo: 

Standard Parallel: 29.500000 Standard Parallel: 45.500000 Longitude of Central Meridian: -96.000000 Latitude of Projection Origin: 23.000000 False Easting: 0.000000 False Northing: 0.000000 

Planar Coordinate Information Planar Distance Units: meters Coordinate Encoding Method: coordinate pair Coordinate Representation Abscissa Resolution: 0.000095 Ordinate Resolution: 0.000095 Geodetic Model Horizontal Datum Name: North American Datum of 1983 Ellipsoid Name: Geodetic Reference System 80 Semi-major Axis: 6378137.000000 Denominator of Flattening Ratio: 298.257222 

(3) Create subregion water body classes from NHD 

In ArcGIS 9.0 python script, the subregion static water body class was defined using the coverage FCODE field: gp.Select_analysis(nhd_alb83_wb, WBstatic, "(FCODE = 39000 OR FCODE = 39002 OR FCODE = 39004 OR FCODE = 43600 OR FCODE = 43601 OR FCODE = 43614 OR FCODE = 43615 OR FCODE = 43616 OR FCODE = 43617 OR FCODE = 43618 OR FCODE = 43619 OR FCODE = 43620) ") 

Small water bodies were extracted using the area field corresponding to equal or less than 5 acres (20,234.3 square meters) in size: gp.Select_analysis(WBstatic,static_sm,"AREA <= 20234.3") Large water bodies were extracted using the area field corresponding to more than 5 acres (20,234.3 square meters) in size: gp.Select_analysis(WBstatic,static_sm,"AREA > 20234.3") Polygon water bodies that are flowing water bodies were extracted from the water body feature class and classified accordingly: gp.Select_analysis(nhd_alb83_wb, WBflowing, "(FCODE = 46000 OR FCODE = 46004 OR FCODE = 46005 OR FCODE = 46006) ") Linear flowing waterbodies were extracted from the nhd drain feature class and classified accordingly: gp.Select_analysis(nhd_alb83_drain, drain_flow, "(FCODE = 46000 OR FCODE = 46004 OR FCODE = 46005 OR FCODE = 46006) ") 

(4) Create subregion NHD margin grid 

The resulting water body class was then exported to a corresponding subregion ArcGIS 9.0 personal geodatabase. The geodatabase water body class was then buffered in a raster environment using the Euclidean Distance tool. An example for one margin distance and one water body class follows: gp.EucDistance_sa(static_sm, EucD_wbwb, "480", "30", Output_direction_raster) 

(5) Reclassify subregion NHD margin grid 

The subregion euclidean distance raster was reclassified into four distance categories: 60m, 120m, 360m, and 480m. gp.Reclassify_sa(EucD_wb, "Value", "0 60 1;60 120 2;120 360 3;360 480 4", Eucwb_rc, "DATA") 

(6) Combine NLCD land cover with NHD margin grid 

The 30m NLCD binary agricultural file (created from a mosaic of rowcrop, fallow, and small grain categories) was combined with the euclidean margin grid to create ag in margin grids. gp.Combine_sa(Eucsm_rc + ";" + us_ag, Comb_wb) 

(7) Create individual subregion grids for water body class margin distances 

The subregion combined grids were then extracted by margin distance into single binary grids for summarization. An example for one margin distance and one water body class follows: # Process: Conditional Margin distance... gp.Con_sa(wb_ag, Input_true_raster_or_constant_value__2_, con_wb120a, "", "Z_z_z2c1 < 3") # Process: Conditional for NLCD Agriculture in Margin... gp.Con_sa(wb_ag, Input_true_raster_or_constant_value, con_wb120, con_wb120a, "Z_z_z2c1 < 3 AND Z_z_z2c2 = 1")

National Hydrography Dataset (NHD) segments were divided into 4 water body classes and each class was buffered to a distance of 60m, 120m, 360m, and 480m in 30m grid format
1.	Ponds/reservoirs: small static water bodies less than or equal to 5 acres in size (up to 2 hectares)
2.	Lakes/reservoirs: large static water bodies greater than 5 acres in size (over 2 hectares)
3.	Rivers/canals: perennial 'polygon' flowing water bodies (i.e., having a left and right bank)
4.	Streams/canals: perennial 'linear' flowing water bodies (i.e., represented as a single center-line)

The margin grids were spatially combined with the agriculture land cover grid (based on the NLCD) to provide the spatial location (and hence total area) of agriculture in each of the margins.  Ratios describing the amount of target crop grown (e.g., corn, cotton) compared to the total area of cropland in each county were created.  This 'crop ratio' was calculated by dividing the NASS 2002 Census of Agriculture acres grown by the acres of 1992 NLCD agriculture for each county.  For example, if County X had 20,000 acres of NLCD agriculture and 10,000 acres of corn according to the Census of Agriculture, then 1 acre of NLCD agriculture in County X is equivalent to 0.5 acres of corn (i.e., a 'corn ratio' of 0.5).  This ratio provides the ability to estimate the amount of target crop for any given amount of NLCD agriculture within a specific county.  It also provides a method to utilize older spatial data (1992 NLCD) to locate agriculture (the only nationally available dataset available), but also use the latest available cropping information (2002 Census of Agriculture) to estimate total area of a given crop.  Utilizing the previously generated 16 'agriculture in margin' grids, the crop ratios were multiplied by the NLCD agriculture acres in the margins to estimate corn and cotton acres in the margins.

Once previously generated margin grids were combined with NLCD agriculture information, the ERF1v2 watersheds (a total of 61,208) were used to summarize agriculture in the four distance margins of small static, large static, perennial polygon flowing, and perennial linear flowing water body classes.  Note that since each water body class was processed separately, and that the same land area can be located within more than one margin (e.g., within the 360m margin of a small pond, and also within the 360m margin of a river), the total area of crop within a specific margin distance for a watershed cannot be calculated by adding the four different water body class results.  The watershed margin results should be evaluated for each water body class.  

Counties and ERF1v2 watersheds were intersected to preserve county identifiers for application of the corn and cotton crop ratios.  Unique County-ERF1v2 codes were used in a zonal statistics tabular summary within the GIS on combined NLCD agriculture in the margins to calculate the amount of NLCD agriculture in each of the margins for each watershed.  Python scripts were used to loop zonal statistics in each subregion.  

Zonal statistic tabular data were loaded into separate water body class geodatabases where the subregion statistics were appended into single agricultural proximity database tables for each water body class / margin distance combination.  Within the database, the crop ratios were applied to the agricultural proximity tables to generate the final corn and cotton margin estimates.

Generate Metadata.