Catalogue of Agriculture Data


In this section, we will catalogue the wide range of different data sources you may find on your farm, or in your farm business. Many of the examples will be taken from grain and oilseed farming but the concepts apply equally to other types of production, such as livestock, horticulture, greenhouse operations, etc. This is not an exhaustive list but will give you a good place to start. Later in the module you will have the chance to actively do an inventory of the key data sources you are using on your farm.

If you take one thing away from this module it should that it pays to get specific when you are thinking and talking about your farm data. A bad habit that many people fall into is referring to data, or agriculture data in general terms only. Although this is easy to do, this limits what you can actually do with your data, and can lead to you missing important risks connected to certain types of data as well.

Don’t fall into this trap. Try as best you can to get specific about the data you are referring to. What type of data is it, how is it being collected, and ultimately what you are trying to do with it.


Ear of Rice Agronomic Data

Agronomic data is about crops and field operations. 

Prescriptions: Variable-rate nutrient, seeding, or crop protection instructions created by agronomy software for use by precision agriculture equipment.

Seeding and Planting: As applied data for seeding rate, planing location, planting date, crop variety, seed type.

Nutrients: As applied fertilizer data including product used and the volume applied.

Crop Protection: As applied crop protection (pesticide, fungicide) data including product used and the volume applied.

Pests and Disease: Field scouting data, pest traps, remote sensing imagery (see below).


Plant Tissue Analysis: Produced by laboratory tests or  bench-top tools .

Grain Moisture: Percentage of water content in grain.

Yield: Volume of production per unit area (e.g. bushels per acre). Yield monitoring is a direct way to assess field production. If harvesting equipment has onboard GPS the location of yield data can be mapped to visualize yield variation across a field. 


How Yield Monitors Work for Grain Harvesting

Yield monitors don’t actually weigh grain. Instead they measure the force against a pressure sensitive plate as the grain moves up the clean grain auger into the grain hopper within the combine. Moisture in the grain can also be measured at the same time by capturing samples of grain in a sensing chamber that is filled periodically to provide a measure of the grain moisture as it changes with time during the harvesting operation. After linking the measured force and moisture measurements with GPS coordinates the measured values are processed using the equipment calibration settings to produce a yield map.


Cow Livestock Data

Data related to the activities, health and wellbeing of livestock.

Animal Identification: Unique identifiers used to track animal movement within the farm and animal conveyances. Tools for animal identification used include RFID tags and RFID readers, either handheld or integrated into a stationary system. GPS ear tags are now being introduced into some operations to monitor location in the field of specific animals. 

Feed Consumption: Consumption, ration contents, feed sources, residual feed intake.

Genetics: Progeny tracking, phenotype selection.

Grazing: Pasture quality, grass or forage biomass. Soil sampling, diversity of forages, the amount of forage, and overall quality can be done through pasture quality assessments.

Barn Conditions: Heating, cooling and ventilation (HVAC).

Health: Drugs and vaccinations, veterinary visits. Tracking the vaccines and antibiotics that have been delivered to animals and other veterinary visits due to sickness is important for monitoring animal health, traceability across the value-chain and for calculating cost of production. 

There are a number of platforms currently in existence that are being more widely adopted that can help automate livestock data collection including   HerdTrax  and  AgriWebb . Technologies like  MoonCyst Bolus  and  SmartBell  can be used to collect real-time animal health data using sensors either within the animal or attached to the outside of the animal.


Desert Island Land Data


Field Boundaries: Despite the simplicity of the concept, managing consistent field boundaries remains a challenging issue for digital agriculture. For more on this issue see the recent work done on field boundaries by the open source  AgStack project from the Linux Foundation  and  VardaAg’s global field ID product .

Source:  VardaAg Global Field ID 

Topography: Elevation - heights, depths, and features of the land. 
  
Soil Fertility: Soil testing is used to provide data on soil organic carbon, salinity and other soil attributes of interest. Soil sampling sites are geolocated using GPS to track changes over time. According to Manitoba Agriculture, in 2016, 41% of Manitoba farmers said they perform soil sampling every year.

One soil testing method is soil electrical conductivity (EC) which measures the ability of soil water to carry electrical current. This can provide a measure of soil salinity as well as soil depth and soil moisture in non-saline environments. 

Soil Moisture: Soil probes provide real-time soil moisture and temperature measurements. 

Tillage: Records of tillage operations including tillage type and depth.

Irrigation and Water Management

Biodiversity

Better Decision Making Using Yield Maps, Satellite Imagery and Soil Tests 

Hank, a grain farmer from Manitoba, hires an agronomy company to zone out his fields which he predicts contain three to four zones based on his experience over the years. Uniform seeding and nutrient applications are just not cutting it as different areas of the farm handle fertilizer and moisture differently.

The agronomy company uses yield maps and  NDVI satellite imagery  to show biomass trends from previous growing seasons to predict zones. The company also goes out into his fields to gather soil data using electrical conductivity monitors and hand texturing. The company compares these more direct measurements with the predicted zones from satellite imagery to create the final zone boundaries for Hank’s farm that he can use to make nutrient and seeding decisions for the next growing season. The company also helps Hank to set up in-field weather stations and soil moisture probes to monitor real time soil and climatic conditions to more accurately predict crop growing conditions leading to more timely crop protection product applications.


Notebook Farm Management Data

Farms are businesses at their core so in addition to crop and livestock data farms also generate a significant amount of conventional business data.

Accounting: Costs and expenses, taxes.

Inventory: On-farm grain storage, seed, crop protection products.

Human Resources: Employment records, recruiting, task assignments.
Marketing: Market data, sales data, customer relationship data.
Regulation: Reporting and compliance data.

Legal: Contracts, leases.


Tractor Machine and Equipment Data

Data related to the operation and maintenance of agricultural machinery and other equipment. Machinery is also used to generate significant quantities of agronomic data, or even livestock data.

Operations: Fuel consumption, load, speed, paths, routes driven over fields.

Maintenance: Maintenance activity, diagnostics, logs, part recalls, downtime.




Sun Behind Rain Cloud Climate and Weather Data

Weather is the most important variable for crop health and closely monitoring weather conditions is essential to determining planting times, crop protection treatment plans, irrigation schedule, and more. 

Weather data can include temperature, wind speed, precipitation, air quality, humidity, solar radiation and atmospheric pressure.   

In-field weather station at  EMILI's Innovation Farms 

Using Local Weather Data for Better Crop Development Predictions

Freely available weather data published by  Environment Canada  is typically only available at the closest town or airport. The same can be true for similar public weather information systems such as the  agriculture weather service from the Province of Manitoba , which provides detailed weather information at 100 additional sites in Southern Manitoba.

Local weather conditions can vary significantly, even over a few kilometers, so the actual weather on-farm is almost always different to some degree. And these small differences in temperature, humidity or rainfall can have big impacts on crop development.

The impact of these differences in local weather conditions can be seen in the image below which shows the difference in crop development predicted using data taken from the regional Environment Canada station and an in-field weather station.  As you can see by the end of the growing season the result is a six day difference in the recommended harvest date.



Remote Sensing: A Multipurpose Tool


Remote sensing technology allows for a wide range of agriculture data to be gathered by unmanned aerial vehicles (UAVs or drones), higher flying aircraft or satellites. Depending on the wavelength of light captured by the onboard cameras and the way the images are processed, different types of agricultural data can be produced:

  • Agronomic: Plant health, disease pressure, weed pressure, crop development, hail damage, crop identification, yield prediction.
  • Livestock: Animal identification, pasture management.
  • Land: Soil characteristics, tillage, cover crops, excess moisture or deficiency.
 
Satellite and aerial imagery is often represented as a two-dimensional map so it can be understood by the grower, a trusted advisor or service provider. For more information on the types of maps available from remote sensing:
  •  Swat Maps:  All Maps tell a Different Story - What is Yours Telling You 
  •  Country Guide:  Field Imagery More than Just a Pretty Picture 


 NDVI and Crop Development

Normalized Difference Vegetation Index (NDVI) is a technique that uses the difference between the red and near infrared light captured by a multispectral camera to monitor vegetation conditions. NDVI works because vegetation absorbs red light and reflects light from the near-infrared band. NDVI imagery is used in agriculture, forestry, or any other application that requires an estimation of biomass. 



Next:  ❌⁠⁠Challenges⁠