Digital Agriculture Words & Definitions

An agtech glossary to help level the playing field.

Check out popular industry words and terms as we seek to demystify some of the jargon and technical terminology used in digital agriculture.

Jump to...

A C D F G H I M P R S V W Y

A

Analysis ReportAn analysis report uses data to answer strategic questions about a farm’s operations, searching for trends and commonalities that can reveal insightful information. For instance, an analysis report might show that while historical recommendations have been to plant soybeans after corn, moving soybean planting dates up can increase yield substantially.

AnalyticsFarmers have always applied “analytics” (aka as common sense!) to their farming decisions. “Data” analytics uses machine learning to analyze a much more comprehensive and multi-layered set of data than a human mind can do on its own. For instance, it might be obvious to a farmer that a particular hybrid performed better than another. But analytics can analyze projected yields, fertilizer needs and seed cost to determine if better yields equal more profit. Also known as “Data Analysis.”

As-Applied DataA record, or map, recording task-specific information as it was executed and where it was executed. Also referred to as “As-Applied Map.” Typically created with GPS technology, data collected is geotagged. As-applied data might include fertilizer rates per application rates and the date they were applied geotagged with exactly where the application was made. This information can be later be used in reporting (see “Analysis Report”).

Artificial IntelligenceAlso referred to as “AI” but not to be confused with Artificial Insemination, also known as “AI!” Artificial Intelligence is a type of computer science using machine learning and deep learning allowing computers to mimic the perception, problem-solving and decision-making capabilities of the human mind. In agriculture, AI learning gives farm management software the ability to analyze thousands of data points, including hybrid varietals, weather conditions, soil type and nutrient levels, to create recommendations per field and zones within a field. See “Machine learning.”


C

CAN-BusA high-speed, hard-wired data network connecting electronic devices. Introduced in 1986, CAN-Bus is used in tractors, combines and other powered vehicles. CAN-Bus creates a network connecting multiple electronic sensors to a controller, which can then be linked to relays or other devices. CAN-bus technology is what allows “precision” machinery to be so precise, helping farmers improve field efficiency with better vehicle diagnostics and by enabling advanced implement management. CAN-bus can also be used to capture high-precision logistical data, which can then be further analyzed. For instance, logging of corn stover biomass production over thousands of acres during harvest, combined with bale drop, pressure and length data, can be used to create yield maps of harvested biomass.


D

Data ActivationThe process of taking collected and stored data and putting it into action to unlock value for the farmer. In farming today, data is logged in most tractors, combines and even equipment — like variable-rate seeders. “Data activation” is the process of transferring data points from many different collection points into a Farm Management Software application where it can be analyzed for valuable trends and insights, like which area of a field yielded best dependent on seeding rates. See “Farm Management Software.”

Data AggregationThe act of bringing together data from a wide range of sources for more powerful insight. In agriculture, aggregated data is compelling because the more information we gathered from many farms on specific products, applications or processes, the more accurate understanding is gleaned. Data collected from one farm can have value. Still, when aggregated and compared with many other farms within a dataset, it can provide a much bigger picture for decision-making. For instance, aggregating data comparing the influence of specific soil types on the growth habits and yield of particular hybrids for not just one farm but 100s of farms with similar soil types in the Midwest. Individual farm privacy is maintained by anonymizing the data points that come from each farm. See “Data Privacy.”

Data AnalysisSee “Analytics.”

Data CompatibilityIntegrating multiple data sets from different sources requires that they be compatible. Data compatibility means data collected on your combine yield monitor can be integrated with data collected from other sources, like a precision seeder, satellite imagery or moisture sensors. In agriculture, data compatibility has been a problem if growers use different platforms for data collection, all with proprietary programs. As digital ag matures, data compatibility is becoming increasingly important to the farmer experience and the industry is moving to unlocked data sources. See “ISOBUS” for more information.

Data IntegrationThe process of merging data from many sources into a unified view. Different types of data integration models create the logistical structure to integrate data, typically involving homogenizing and normalizing data so it can be combined into a master server. In agriculture, this would be taking compatible data (see data compatibility) from various sources. Sources might include the internet (commodity market trends), sensors, satellites, weather stations, farm equipment, government databases, agribusiness databases and farmer-generated data itself and aggregated that data into one master view.

Data LayersA data layer takes complex data from many sources. It creates and stores it into an organized, accurate, standardized set of data used by multiple systems. A data layer can be specific to one farm’s data or a broader set of data, like the USDA Cropland Data Layer that releases an annual georeferenced, crop-specific land cover map using satellite imagery. Every farm operation is highly complex and standardized data layers from many sources give more accurate insight into farm decision-making.

Data PrivacyThe governance of how data is collected, shared and used. There is currently no dedicated U.S. data privacy law regulating any sector (including agriculture), although some states have data privacy laws that may impact agricultural data collection. Data privacy is self-regulated voluntarily by ag tech providers. The prime components of data privacy include “anonymizing” collected data, aka removing the individual farmer’s information from the data, and then storing and aggregating the data into datasets. This allows providers to collect information that empowers more accurate systems while protecting individual producer data.

Data VisualizationDisplaying information in a visual format to understand the context and communicate the underlying meaning. Data visualization in digital ag is usually in a pie chart, graph, table, diagram or illustration. Data visualization is helpful in quickly identify trends, predict outcomes and identify areas that need attention or improvement. For instance, a graph might be used to visualize nitrogen prescription levels and the following yields based on application timing, amounts or hybrids planted.

Digital AgOr, “Digital Agriculture.” The integration of digital, “Big Data” technologies in crop and livestock management and other processes in agriculture. Digital ag uses digitized data science techniques to collect, sort and analyze data collected in agricultural production. Using AI (see “Artificial Intelligence”) and deep machine learning (see “Machine Learning”), digital ag enables predictive and analytical capabilities beyond what is capable by a human mind. Digital ag offers farmers the ability to gather more precise information, become more efficient in their operational practices, and generate deep analytical models to help farmers evaluate new strategies before taking the actual risk.

Directed ScoutingIn-season reports directing field scouting via GPS-enabled, site-specific field hot spots identified by high-resolution satellite imagery (see “Satellite Imagery”). Directed scouting uses vegetation index maps (see definition below) to identify areas of high scouting priority that aren’t growing as expected. A directed scouting report, usually created weekly, tells a farmer exactly where to go in their fields to ground-truth a potential problem. Directed scouting might identify an area of poor germination early enough in the season that the section can be replanted and still yield well, for instance. Directed scouted uses GPS-enabled mapping features that can be accessed on a smartphone or imagery when WiFi or cell phone reception is disabled. See also “Scouting Maps.”


F

Farm DataThe information collected by individual farms through various means, including machine-enabled data sources such as combine yield monitors and variable-rate seeders or specific digital services, such as moisture sensors, satellite imagery or drone imagery. Farm-specific data can be layered with other data, such as regionally collected agronomic data, for more insights. To be useful, farm data must be activated (see “Data Activation”) using a software platform (see “Farm Management Software”).

Farm Management Software (FMS)Software helping farmers manage and optimize their operations, including everything from daily work orders, farm financials, agronomic prescriptions and real-time, in-season decision making. FMS programs take data collected from universal data sets (like USDA-generated crop information or soil maps) and aggregates it with on-farm generated data, like yield data or soil tests, for insightful analysis. FMS is typically customized in approach. For instance, software specific to livestock producers versus field crop production. Depending on the software provider, FMS may have varying products that can be mixed and matched to achieve the best level of service per customer.

Field BoundariesThe creation of field boundaries allows for fields to be grouped and sorted and is the starting point for working with digital agronomic services. Once boundaries are set — most Farm Management Software (FMS) systems (see above) automatically create field boundaries when planting or harvest data is uploaded. Crops planted, inputs applied and other services such as satellite imagery can be ordered, organized and utilized. And data collected from those fields can be accurately collected into a map layer (see “Map Layers”) for FMS analysis.

Field ProfitabilityAlso referred to as “field-level profitability.” The ability of a Farm Management Software (FMS) program to create a per-field analysis of profitability. Rather than looking at whole-farm profit or loss, FMS can identify profitability by field, allowing farmers to make specific field decisions to increase overall farm profitability. This may include deciding to drop leases on unprofitable fields or change input levels; crops planted or other farm management practices.

Field LevelThe ability to identify and track the effect of agronomic and operational decisions on a field level. Although farmers have always known that each field is unique, they have not traditionally been able to evaluate decision-making at a field level. They have usually made whole-farm, cover-the-bases operational decisions rather than customizing practices per field. By setting up field boundaries (see above) and collecting data that is analyzed per field, farmers can then make management decisions using precision ag planting and product application machinery to customize for best profitability per field (see also “Field Profitability.”)


G

GeolocationSometimes called “geotagging.” The geographical location (latitudinal and longitudinal) of an internet-connected device. Instead of using GPS to locate a specific spot (GPS coverage is sometimes spotty), geolocation uses WiFi routers, mobile phone towers and beacons to pinpoint an exact location. Geolocation is helpful in direct scouting (see “Directed Scouting”) applications, telling farmers exactly where to go in a field to ground truth a potential seasonal trouble spot identified by satellite imagery. When real-time geolocation services don’t work (no cell phone signal, for instance), site-specific images located within a map can be used instead.

GISGeographic information system. Connects data points to maps integrating location data with descriptive data. GIS uses GPS (a U.S based satellite system that finds the exact location of something on the Earths’ surface) and other means of recording geographic location (see “Geolocation” above) to create geographically-integrated databases. GIS allows for complex spatial analyses in precision agriculture applications, such as layering data like soil type, wind direction, rainfall, topography and elevation in a field and even exact positions within a field. GIS technology greatly enabled precision agriculture solutions.


H

High-Frequency Satellite ImageryHigh-resolution images taken near-daily by satellites orbiting the Earth at a relatively low altitude. High-frequency satellite imagery maps three-dimensional disparities in crop and soil conditions, identifying “hot spots” where stressed plants are developing. Because of an ever-expanding number of surveillance satellites, the quality and quantity of data collected by satellite imagery enable a quick response to potential trouble spots as they are developing. This allows farmers to head off in-season issues before they detrimentally affect harvest yields. In addition, archiving satellite images and then comparing satellite data collected throughout a season (or one season versus another) can be used to spot trends and identify potential changes in operational practices.


I

IoTAcronym for “Internet of Things, referring to the system of the internet (cloud-based), connected objects that collect and transfer data over a wireless network without human intervention. IoT in action would include activities like automatic uploads of daily commodity market reports into farm-specific financial reporting in a farm management software system. Or automatic downloads of satellite imagery into directive scouting apps would alert farmers to field trouble spots as they occur in season.

ISOBUSA communication protocol connecting tractors, software and equipment of major manufacturers. ISOBUS, introduced in 2001, solves the problem of “proprietary” data collection programs, which was common when digital ag was first developing and allowed exchanging information with a universal language between equipment. Most major agricultural manufacturers have adopted and implemented ISOBUS or are in the process of sunsetting proprietary programs and transferring them to ISOBUS communication systems. For farmers, this means it doesn’t matter what brand their driving or using, their equipment’s data-collecting devices should be using the same communication protocols.


M

Machine DataAlso called “machine-generated data.” Information automatically created using networked devices such as computers and mobile phones. In agriculture, machine data is also created (literally) by farmer’s machines through the use of data-collecting hardware (see “CAN-Bus”). This hardware is mounted into tractor and combine cabs and implements used in precision ag practices such as seeders, balers and fertilizer spreaders and guidance systems (autosteer, GPS-enabled tractors, etc.)

Machine LearningA branch of artificial intelligence (AI) that creates applications that learn from data, improving accuracy over time without being re-programmed. Machine learning is all around us, like when we log into our Netflix account and are recommended new movies based on past viewings. In farming, machine learning technology is being applied in many ways. Farm robots learn to identify and eliminate weed species. In yield prediction, machine learning looks at historical data, weather, economic conditions and many other factors to create crop yield mapping and estimations used by farmers when making crop management decisions. See “artificial intelligence” for more information.

Map LayersA map layer is a specific set of data pinpointed by field boundaries (see “field boundaries” above). Map layers are collected from many different data collection points like soil maps, precision planters, combine yield maps and satellite imagery, but are all layered into the same specific map. A farmer can then compare map layers for further insight into that particular field, for instance, target seeding rates as applied versus yield.

ModellingPresenting quantitative knowledge about an agricultural process to define priorities and glean a better understanding. Modelling techniques are often used in “crop modelling,” helping to understand better crop growth interactions with external factors like soil, weather, and fertility. Models can be used to simulate crop development and yield based on changes in external factors. For instance, modelling might be used to determine whether another application of nitrogen is needed on a corn planting to reach optimum yield after an unexpected cold snap or heavy rain event. (See “Nitrogen modelling.”)

Management ZoneA sub-region of a field that has similar factors limiting yield potential. For instance, the ridge of a field exposed to more erosion, a low spot that might be wet in the spring or a different soil profile through a field section. Each zone responds differently. By mapping and labelling management zone fertility levels, seeding rates and other factors, in-season practices can be adjusted and accommodated to optimize yield.


N

Nitrogen ModellingA computer program that estimates the nitrogen cycle and is customized to the field (and zone within the field) and per crop. Estimating nitrogen availability through a growing season is something agronomists and crop consultants have been doing for years. What is different with digital nitrogen modelling is the ability to analyze multiple variables (including seasonal weather events) and then create prescriptions (see “prescription” definition below) to encompass an entire season’s nitrogen use. Nitrogen modelling allows farmers to get extremely specific in their nitrogen applications, optimizing crop yields while limiting input costs and decreasing nitrogen run-off.


P

Precision FarmingAlso known as “precision agriculture” or “precision ag.” A farming management concept based on observing, measuring, and then precisely responding to variability within crop production management practices to maximize yield and profitability. Precision agriculture involves gathering data, creating customized prescriptions (see “prescriptions” for more information) and then ensuring the accuracy of operations using GPS and GIS-based (see “GIS”) operating systems in precision ag-enabled equipment.

PrescriptionSometimes called “prescription farming.” Variable rates of seeding and inputs based on per field and management zone (see definition above) data obtained through yield monitors, soil maps and sampling, historical data and remote sensing, and supplier data like hybrid seed variability. Farmers work with their crop consultants for a “per field” prescription based on the hybrid they are planting and their field data. The prescription is then used to inform precision ag technology (like variable-rate seeders) using GPS-location to plant seeds or apply inputs at the precisely correct amount at the exact right spot within a field.


R

Remote SensingThe act of detecting and identifying an object, series of objects, or landscape without having the sensor in direct contact with the object. Remotely sensed images — typically taken by satellite, drone or aircraft images — can assess field conditions without physically being in the field and from a viewpoint high above the field. Using visible wavelengths a human eye can see plus wavelengths humans can’t, remote sensing observes the leaf color to assess overall plant condition. Farmers use remote sensing to identify nutrient and water deficiencies, disease and pest outbreaks, plant populations and herbicide and wind damage. Information collected from remote sensing can also be used as base maps in creating variable rate applications of fertilizers and pesticides.


S

Satellite Imagery ‐ See High-Frequency Satellite Imagery.

Scouting MapsVarious map layers using different wavelengths captured in remote sensing imagery (see “Remote Sensing” above) identify issues within a mapped area. For instance, one scouting map might analyze vigor, biomass development and basic crop health while another, looking at different wavelengths, analyzes chlorophyll content to determine nitrogen recommendations. Or another scouting map might look at spectral bands indicating areas affected by water stress.

Side-by-Side MappingThe ability to visualize different sets of data on a map of the same field, useful for spotting trends, ground-truthing fertility applications and tracking crop growth. For instance, a farmer can compare satellite or drone images side-by-side weekly to track the progress of a crop and target harvest dates. Or, a farmer can compare a map of nitrogen prescription rates as applied to a map of vegetative growth in the same field to track the crop's response to nitrogen application.

SSURGOAcronym for Soil Survey Geographic Database. A digital version of the soil maps maintained by the Natural Resources Conservation Service (NRCS). SSURGO maps include soil types and their distribution, including soil characteristics and soil properties and address risks and suitability for various agricultural uses.


V

Variable Rate Technology (VRT)The use of GPS and precision placement technology to guide the application of products to specific locations with a field. VRT uses machines working with systems that use sensors, controls and machinery to vary the rate of applications of inputs (like fertilizers or pesticides) or seeds based on specific locations and conditions within a field. VRT uses an application guidance map to direct application rates. VRT application maps can also be layered with other field-specific information (like harvest yields) to create a deep analysis of farm agronomic practices.

Vegetation MapsAlso known as vegetative index or vegetation index maps. Using different spectral bands captured via remote sensing (see above) to visualize different types of plant stress in a specific field to identify developing field problems. See also “Scouting Maps.”


W

Whole-Farm DigitizationThe ability to combine agronomic, practice-based prescriptions with by-field profitability insights and whole-farm financial goals to make by-field decisions using comprehensive farm operational understanding. Whole farm digitization is the natural progression of digital ag (see definition above) applications, enabled by farm management software (definition above).


Y

Yield CalibrationProcedures to calibrate a yield monitor for specific conditions such as grain type, flow and moisture to ensure accurate records of harvest yields. Calibration converts yield sensors to physical parameters for precise measurements. To achieve accurate yield monitor data, producers should follow manufacturer’s guides for calibrating yield monitoring systems.

Yield DataData generated by a yield monitor (see definition below) on harvesting equipment (typically a combine). Yield data typically includes the volume of grain harvested, the variability of moisture in the grain as harvested, the clean grain elevator speed sensor (to improve the accuracy of grain flow measurements), GPS signals (used in yield mapping, see definition below), combine header position and combine speed during logged intervals. Yield data can be used to calculate final harvest and spot trends. Yield data can also be layered with other field data maps (like variable-rate seeding or per-field nitrogen prescriptions) for better insight into per-field decision making.

Yield MapsData maps created during harvest via a combine yield monitor (definition below) using GPS to collect georeferenced data on crop yield and characteristics, such as moisture content. Various sensors are used in mapping crop yields and each map is tagged with latitude and longitude coordinates.

Yield MonitorA device installed on harvest machines gathering vast amounts of information measuring yields and yield data (see “yield data” above). Yield monitors measure grain flow, moisture and other parameters (such as GPS location) related to field productivity. Yield monitors give farmers useful information helping them to assess things such as when to harvest plus the effects of fertilizer applications, seed choices and weather on final yields and harvest quality.