The Moderate Resolution Imaging Spectroradiometer (MODIS) Irrigated Agriculture Datasets for the Conterminous United States (MIrAD-US)

The Moderate Resolution Imaging Spectroradiometer (MODIS) Irrigated Agriculture Datasets for the Conterminous United States (MIrAD-US)

MIrAD: An Overview

MIrAD is the Irrigated Agriculture Datasets for the Conterminous United States (MIrAD-US). It was funded by the USGS National Water-Quality Assessment (NAWQA) program. This product maps the irrigated agricultural land across the conterminous United States at 250-m and 1-km spatial resolution. Version 3 of the MIrAD was released on April 16th, 2015 and had the data for the years 2002, 2007, and 2012. USGS EROS Center at Sioux Falls, SD plans the product update for every 5 years, synchronized with the update of the Census of Agriculture by the U.S Department of Agriculture (USDA) but contingent upon availability of eMODIS data and funding. The MIrAD datasets are open for public and can be found at USGS website.


The goal of the MIrAD project is to provide the USGS and other researchers with complete and consistent public domain information on the spatial distribution of irrigated lands across the conterminous United States.

Making of MIrAD Dataset

The MIrAD primarily uses three datasets: eMODID time series data, National Land Cover Database (NLCD) and the county irrigated area statistics were obtained from the 2012 USDA Census of Agriculture (Brown and Pervez 2014; Pervez and Brown 2010).

The eMODIS time-series data set used in this research was created at the USGS EROS Center through funding from the USGS Land Remote Sensing Program. Data are made avaialble through the Earth Explorer site.
The NLCD was produced through a cooperative project conducted by the Multi-Resolution Land Characteristics (MRLC) Consortium. The MRLC Consortium is a partnership of federal agencies, consisting of the U.S. Geological Survey (USGS), the National Oceanic and Atmospheric Administration (NOAA), the U.S. Environmental Protection Agency (EPA), the U.S. Department of Agriculture (USDA), the U.S. Forest Service (USFS), the National Park Service (NPS), the U.S. Fish and Wildlife Service (FWS), the Bureau of Land Management (BLM), and the USDA Natural Resources Conservation Service (NRCS). The database is available online.
The county irrigated area statistics were obtained from the USDA Census of Agriculture. The Census of Agriculture, taken every 5 years, is a complete count of U.S. farms and ranches and the people who operate them. The Census looks at land use and ownership, operator characteristics, production practices, income and expenditures, and many other areas. The Census was conducted by the National Agricultural Statics Service (NASS) and is available online.


The overall concepts guiding the MIrAD-US mapping approach are as follows:
a. Irrigated crops should have higher peak Normalized Difference Vegetation Index (NDVI) values than non-irrigated crops. Please refer to MIrAD Metadata and/or Brown and Pervez 2014; Pervez and Brown 2010 papers for detailed information.

b. The spectral values (and consequently, the NDVI) of irrigated and non-irrigated crops ought to be similar in seasons with optimum precipitation.  Severe drought conditions will maximize the difference between irrigated and non-irrigated crops.  The annual period with the most severe and widespread drought conditions should be selected for the model.

c. The growing season peak NDVI, whenever it occurs, is preferred as opposed to choosing a specific time period, because the peak greenness time varies for each crop and for each geographic region of the United States.

Figure 1: Schematic diagram of the Moderate Resolution Imaging Spectroradiometer (MODIS) Irrigated Agriculture Dataset (MIrAD-US) methodology. (Pervez, M.S. and Brown, J.F., 2010)

Process Steps

The MIrAD-US model was developed and applied in ARC GIS 10.x and Python scripting based on three input datasets (please refer to MIrAD Metadata and/or Brown and Pervez 2014; Pervez and Brown 2010 papers for detailed information):
(1) 250-m spatial resolution eMODIS annual maximum vegetation index data,
(2) county irrigation summary statistics of the U.S. Department of Agriculture (USDA) Census of Agriculture, and
(3) an agriculture land cover mask based on the 2011 NLCD. The model was run for each individual county and the results merged for all counties.

The model steps were implemented as follows:
a. An agriculture land cover mask was derived from the NLCD; only 250-m cells categorized as dominated by class 81 [Pasture/Hay] and 82 [Cultivated Crops] were included from the mask application.
b. The annual peak eMODIS NDVI values and corresponding cell counts were extracted for the agriculture lands only using agriculture land cover mask.
c. Annual peak NDVI values and related cell counts were sorted in descending order
d. The total area was calculated for each peak NDVI value and then accumulated in descending order.
e. The accumulated area for each peak NDVI was compared with the irrigated acreage reported by the USDA Census of Agriculture.
f. When the total accumulated area computed for the peak NDVI cells was closest to, and just barely greater than, the area reported by the USDA Census of Agriculture, then that peak NDVI value from the descending list was selected as the peak NDVI threshold used to identify irrigated cells.
g. In a final post-processing step, all single 250-m irrigated pixels are filtered from the irrigated map based on the assumption that most irrigation in the United States is found in fields which are larger than 250 m. Therefore, this methodology may not capture very small, isolated irrigated fields.

Figure 3: National change in irrigated agriculture from 2002 to 2007 (Brown, J.F. and Pervez, M.S., 2014).


Although these data have been processed successfully on a computer system at the USGS, no warranty expressed or implied is made by the USGS regarding the use of the data on any other system, nor does the act of distribution constitute any such warranty.  Data may have been compiled from various outside sources.  Spatial information may not meet National Map Accuracy Standards.  This information may be updated without notification. The USGS shall not be liable for any activity involving these data, installation, fitness of the data for a particular purpose, its use, or analyses results.


Brown, J.F. and Pervez, M.S., 2014, Merging remote sensing data and national agricultural statistics to model change in irrigated agriculture, Agricultural Systems, 127; doi:10.1016/j.agsy.2014.01.004.

Pervez, M.S. and Brown, J.F., 2010, Mapping irrigated lands at 250-m scale by merging MODIS data and national agricultural statistics, Remote Sensing, 2(10), 2388-2412; doi:10.3390/rs2102388. [Available online at]

1 thought on “The Moderate Resolution Imaging Spectroradiometer (MODIS) Irrigated Agriculture Datasets for the Conterminous United States (MIrAD-US)”

  1. Pingback: Mapping Center-pivot Irrigation using Google Earth Engine – DINESH SHRESTHA

Leave a Reply

%d bloggers like this: