Machine learning in agricultural planting, growing, and harvesting contexts
A crop prediction system performs various machine learning operations to predict crop production and to identify a set of farming operations that, if performed, optimize crop production. The crop prediction system uses crop prediction models trained using various machine learning operations based on geographic and agronomic information. Responsive to receiving a request from a grower, the crop prediction system can access information representation of a portion of land corresponding to the request, such as the location of the land and corresponding weather conditions and soil composition. The crop prediction system applies one or more crop prediction models to the access information to predict a crop production and identify an optimized set of farming operations for the grower to perform.
1. A method comprising:
selecting a portion of land from a cluster of identified portions of land, the selected portion of land associated with crop growth information describing 1) characteristics of the selected portion of land, 2) a first set of farming operations to be performed, and 3) a first expected crop productivity corresponding to the first set of farming operations;
applying a prediction model trained on historical crop growth information from a plurality of geographically diverse locations to inputs comprising the characteristics of the selected portion of land, the first set of farming operations to be performed, and the first expected crop productivity corresponding to the first set of farming operations, wherein the prediction model optimizes a second expected crop productivity by producing an output comprising a second set of farming operations that can produce the optimized second expected crop productivity, wherein the second set of farming operations comprises an agricultural chemical to apply; and
for the selected portion of land, 1) modifying the first set of farming operations to be performed based on the identified second set of farming operations, and 2) modifying a user interface displayed by a client device of the user to display a crop growth program based on the modified first set of farming operations.
2. The method of claim 1 , further comprising normalizing the crop growth information associated with the selected portion of land into a format or scale based on the crop growth information used to train the prediction model.
3. The method of claim 2 , wherein normalizing the crop growth information associated with the selected portion of land comprises one or more of: removing format-specific content from the crop growth information, removing or modifying portions of the crop growth information associated with values that fall outside of one or more predefined ranges, and scaling image information such that each image pixel represents a same distance.
4. The method of claim 1 , wherein a first set of farming operations identifies one or more of: a type or variant of crop to plant, an intercrop to plant, a cover crop to plant, a date to plant a crop, a planting rate, a planting depth, a microbial composition, a date to apply a microbial composition, a rate of application for a microbial composition, an agricultural chemical to apply, a date to apply an agricultural chemical, a rate of application for an agricultural chemical, type of irrigation, a date to apply irrigation, a rate of application for irrigation, whether to replant the crop, whether to replant a different crop within the portion of land, a replant date, a type of nutrient to apply, a quantity of nutrient to apply, a location to apply a nutrient, a date to apply a nutrient, a frequency to apply a nutrient, a nutrient application method, a quantity of water to apply, a type of treatment to apply, a quantity of treatment to apply, a location to apply treatment, a date to apply treatment, a frequency to apply treatment, a treatment application method, a harvest date, a harvest method, a harvest order, a piece of equipment to use or purchase, a drainage method to implement, a crop insurance policy to purchase, a period to store a crop, one or more potential crop brokers, one or more potential crop purchasers, one or more harvested crop purchase prices, and one or more harvested crop qualities, wherein the harvested crop qualities includes at least one of: a crop moisture content, a crop protein content, a crop carbohydrate content, a crop oil content, a crop fat content, a crop color, a crop hardness, a measure of wet gluten, a number or percentage of broken grains, a toxin level, a damage level, whether the crop is organic, whether the crop is shade grown, whether the crop is greenhouse grown, whether the crop is fair-wage grown, whether the crop is no-till grown, when the crop is pollution-free grown, when the crop is carbon neutral, and a grading or certification by an organization or agency.
5. The method of claim 1 , wherein portions of land in the cluster of identified portions of land are identified based on having a threshold similarity to one another, and wherein the threshold similarity comprises a measure of similarly based on one or more of: geography, climate, soil type, soil composition, soil and atmospheric temperature, number of growing degree days, and precipitation.
6. The method of claim 1 , wherein the crop growth program is periodically modified in response to re-applying the prediction model to updated crop growth information associated with the selected portion of land.
7. The method of claim 1 , wherein the portions of land are fields, plots of land, planting regions, zones, management zones, or sub-portions thereof.
8. The method of claim 1 , wherein the prediction model is applied to the crop growth information associated with the selected portion of land in response to a triggering event, wherein the triggering event comprises one of: a weather event, a temperature event, a plant growth stage event, a water event, a pest event, a fertilizing event, a farming machinery-related event, a market event, a contract event, and a product supply event.
9. The method of claim 1 , wherein the cluster of portions of land are selected from locations associated with one or more of: a threshold geographic diversity, a threshold environmental diversity, a threshold geographic similarity, and a threshold environmental similarity.
10. The method of claim 1 , wherein the prediction model is applied to the crop growth information associated with the selected portion of land in response to a request from a grower, a technology provider, a service provider, a commodity trader, a broker, an insurance provider, an agronomist, or other entity associated with one or more portions of land in the cluster of portions of land.
11. The method of claim 1 , wherein the crop growth information associated with the selected portion of land further describes one or more of: rainfall associated with the selected portion of land, canopy temperature associated with the selected portion of land, soil temperature of the selected portion of land, soil moisture of the selected portion of land, soil nutrients within the selected portion of land, soil type of the selected portion of land, topography within the selected portion of land, humidity associated with the selected portion of land, growing degree days associated with the selected portion of land, microbial community associated with the selected portion of land, pathogen presence associated with the selected portion of land, prior farming operations performed at the selected portion of land, prior crops grown at the selected portion of land, other historical field information associated with the selected portion of land, a crop plant stage, a crop color, a crop stand count, a crop height, a crop root length, a crop root architecture, a crop immune response, a crop flowering, and a crop tasseling.
12. The method of claim 1 , wherein the prediction model comprises one or more of: a generalized linear model, a generalized additive model, a non-parametric regression operation, a random forest classifier, a spatial regression operation, a Bayesian regression model, a time series analysis, a Bayesian network, a Gaussian network, a decision tree learning operation, an artificial neural network, a recurrent neural network, a reinforcement learning operation, linear/non-linear regression operations, a support vector machine, a clustering operation, and a genetic algorithm operation.
13. The method of claim 1 , wherein the crop growth information associated with the selected portion of land is collected from one or more of: sensors located at the selected portion of land, satellites, aircraft, unmanned aerial vehicles, land-based vehicles, and land-based camera systems.
14. The method of claim 1 , wherein modifying the first set of farming operations comprises replacing a first crop to be planted with a carbon neutral plant identified by the second set of farming operations.
15. The method of claim 1 , wherein the prediction model is trained on crop growth information accessed from sensors located at the plurality of geographically diverse locations.
16. The method of claim 1 , wherein the prediction model is trained on crop growth information captured at different times during one or more growing seasons.
17. The method of claim 1 , wherein the crop growth information associated with the selected portion of land comprises carbon dioxide content information.
18. The method of claim 1 , wherein the prediction model is trained on crop growth information comprising carbon dioxide content information.
19. A system comprising:
a processor; and
a non-transitory computer-readable storage medium storing executable instructions that, when executed by the processor, cause the processor to perform steps comprising:
selecting a portion of land from a cluster of identified portions of land, the selected portion of land associated with crop growth information describing 1) characteristics of the selected portion of land, 2) a first set of farming operations to be performed, and 3) a first expected crop productivity corresponding to the first set of farming operations;
applying a prediction model trained on historical crop growth information from a plurality of geographically diverse locations to inputs comprising the characteristics of the selected portion of land, the first set of farming operations to be performed, and the first expected crop productivity corresponding to the first set of farming operations, wherein the prediction model optimizes a second expected crop productivity by producing an output comprising a second set of farming operations that can produce the optimized second expected crop productivity, wherein the second set of farming operations comprises an agricultural chemical to apply
for the selected portion of land, 1) modifying the first set of farming operations to be performed based on the identified second set of farming operations, and 2) modifying a user interface displayed by a client device of the user to display a crop growth program based on the modified first set of farming operations.
20. A non-transitory computer-readable storage medium storing executable instructions that, when executed by a processor, cause the processor to perform steps comprising:
selecting a portion of land from a cluster of identified portions of land, the selected portion of land associated with crop growth information describing 1) characteristics of the selected portion of land, 2) a first set of farming operations to be performed, and 3) a first expected crop productivity corresponding to the first set of farming operations;
applying a prediction model trained on historical crop growth information from a plurality of geographically diverse locations to inputs comprising the characteristics of the selected portion of land, the first set of farming operations to be performed, and the first expected crop productivity corresponding to the first set of farming operations, wherein the prediction model optimizes a second expected crop productivity by producing an output comprising a second set of farming operations that can produce the optimized second expected crop productivity, wherein the second set of farming operations comprises an agricultural chemical to apply; and
for the selected portion of land, 1) modifying the first set of farming operations to be performed based on the identified second set of farming operations, and 2) modifying a user interface displayed by a client device of the user to display a crop growth program based on the modified first set of farming operations.