18.2 Geospatial Data for Agriculture
Agricultural-related geospatial data, also called Agro-geodata, is a kind of geospatial
data that is defined as the datasets with a spatial component on Earth (Di 2016).
Studies have shown that more than 80% of the datasets in the world are geospatial
data (Hahmann et al. 2011). Therefore, Agro-geodata also has multiple data categories, which derive from both traditional data sources and new data sources for
agriculture modeling, Agro-geodata life cycle ranging from data collection, information processing, and knowledge derivation in the agricultural domain.
18.2.1 Data Categories
18.2.1.1 CDL Data
In 2009, the US Department of Agriculture (USDA) National Agricultural Statistics
Service (NASS) Cropland Data Layer (CDL) program played an important role in
providing timely, accurate, and meaningful statistics services for US agriculture.
This program classifies US crop cover types by using satellite imagery from
Advanced Wide Field Sensor (AWiFS) and Landsat TM 5 and ETM+ 7 (Han
et al. 2012). Specifically, the CDL data derives from USDA Farm Service Agency
(FSA) Common Land Unit (CLU) data, associated Administrative Data 578 attribute
data which includes all crop-specific information such as crop type, status, and
intention codes, NASS’s June Agricultural Survey (JAS) data, and US Geological
Survey (USGS)‘s National Land Cover Datasets (NLCD) as ground truth, provides
the specific crop and other land cover classifications encompassing the Contiguous
United States (CONUS) in a geo-referenced raster dataset (Boryan et al. 2011).
The standard CDL product is disseminated in GeoTIFF format and projected in
the Albers equal-area projection. These valuable products have been published from
1997 to 2016 (continue to update every year) and covered all 48 conterminous states
and the District of Columbia from the crop year of 2008. For the crop year
2006–2009, the spatial resolution of CDL data is 56 m, and for other years, the
spatial resolution is 30 m. Figure 18.1a shows the 2016 CDL data.
The CDL data could deliver approximated acreage estimates by counting pixels
within a given area of interest (AOI) (e.g., an agricultural region); the derivative
products could be downloaded at the state/country/ASD/region level in a compressed form from the NASS website or CropScape (https://nassgeodata.gmu.edu/
CropScape). The CDL data has been used as a significant data source in agricultural
research fields, such as crop rotation (Stern et al. 2012), land cover monitoring
(Hansen and Loveland 2012), drought monitoring (Deng et al. 2013), flood and
progress monitoring (Di et al. 2017), biomass monitoring (Chandola and Vatsavai
2011), and carbon accounting (West et al. 2010).
18 Design and Implementation of Geospatial Data Services for Agriculture
387
Agricultural-related geospatial data, also called Agro-geodata, is a kind of geospatial
data that is defined as the datasets with a spatial component on Earth (Di 2016).
Studies have shown that more than 80% of the datasets in the world are geospatial
data (Hahmann et al. 2011). Therefore, Agro-geodata also has multiple data categories, which derive from both traditional data sources and new data sources for
agriculture modeling, Agro-geodata life cycle ranging from data collection, information processing, and knowledge derivation in the agricultural domain.
18.2.1 Data Categories
18.2.1.1 CDL Data
In 2009, the US Department of Agriculture (USDA) National Agricultural Statistics
Service (NASS) Cropland Data Layer (CDL) program played an important role in
providing timely, accurate, and meaningful statistics services for US agriculture.
This program classifies US crop cover types by using satellite imagery from
Advanced Wide Field Sensor (AWiFS) and Landsat TM 5 and ETM+ 7 (Han
et al. 2012). Specifically, the CDL data derives from USDA Farm Service Agency
(FSA) Common Land Unit (CLU) data, associated Administrative Data 578 attribute
data which includes all crop-specific information such as crop type, status, and
intention codes, NASS’s June Agricultural Survey (JAS) data, and US Geological
Survey (USGS)‘s National Land Cover Datasets (NLCD) as ground truth, provides
the specific crop and other land cover classifications encompassing the Contiguous
United States (CONUS) in a geo-referenced raster dataset (Boryan et al. 2011).
The standard CDL product is disseminated in GeoTIFF format and projected in
the Albers equal-area projection. These valuable products have been published from
1997 to 2016 (continue to update every year) and covered all 48 conterminous states
and the District of Columbia from the crop year of 2008. For the crop year
2006–2009, the spatial resolution of CDL data is 56 m, and for other years, the
spatial resolution is 30 m. Figure 18.1a shows the 2016 CDL data.
The CDL data could deliver approximated acreage estimates by counting pixels
within a given area of interest (AOI) (e.g., an agricultural region); the derivative
products could be downloaded at the state/country/ASD/region level in a compressed form from the NASS website or CropScape (https://nassgeodata.gmu.edu/
CropScape). The CDL data has been used as a significant data source in agricultural
research fields, such as crop rotation (Stern et al. 2012), land cover monitoring
(Hansen and Loveland 2012), drought monitoring (Deng et al. 2013), flood and
progress monitoring (Di et al. 2017), biomass monitoring (Chandola and Vatsavai
2011), and carbon accounting (West et al. 2010).
18 Design and Implementation of Geospatial Data Services for Agriculture
387