Multi-Crop Specific Area Frame Stratification Based on Geospatial Crop Planting Frequency Data Layers

Area sampling frames (ASFs) are the basis of many statistical programs around the world. When an ASF's stratification is based on generalized percent cultivation, the ASF usually cannot identify the planting location of specific crops targeted for agricultural surveys. To improve the accuracy, objectivity and efficiency of crop survey estimates, an automated stratification method based on geospatial crop planting frequency data is proposed. The Crop Planting Frequency Data Layers are crop specific geospatial data sets derived from multi-year Cropland Data Layers. Therefore, the ASF stratification based on the crop planting frequency data is crop specific. This paper investigates using 2008-2013 geospatial Crop Frequency Data Layers to create a novel multi-crop specific stratification for South Dakota, U.S. The crop specific ASF stratification is developed based on crop frequency statistics calculated at the primary sampling unit (PSU) level based on the corn, soybean and wheat planting frequency data layers, three major crops in South Dakota. Strata are formed using a k means clustering algorithm. It is observed that the crop frequency based ASF stratification predicts corn, soybean and wheat planting patterns well as verified by 2014 Farm Service Agency (FSA) Common Land Unit (CLU) and 578 administrative data. This finding demonstrates that the novel multi-crop specific stratification based on crop planting frequency data is crop type independent and applicable to all major crops. Further, these results indicate that the new multi-crop specific ASF stratification has great potential to improve ASF accuracy, efficiency and crop estimates.