Precision crop production-function models
Inventors
Campisi-Pinto, Salvatore • Crowley, David • Lovatt, Carol J.
Assignees
University of California San Diego UCSD • University of California Irvine
Publication Number
US-11109523-B2
Publication Date
2021-09-07
Expiration Date
2037-08-21
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Abstract
Systems, apparatuses, and methods as described herein relate to analyzing plant nutrient-yield relationships and identifying optimum plant fertilization strategies to improve individual plant yields and overall plant yields. Nutrient data and yield data for plants in all yield ranges, rather than nutrient data and yield data for only the best-performing plants, is categorized based on the yield data and then analyzed to determine the yield as a function of nutrient condition for each nutrient element for the plants in each yield category. The nutrient-yield function can be modeled using a polynomial equation of various orders that can be used for determining the nutritional condition to improve crop production for plants in each yield category.
Core Innovation
The invention relates to systems, apparatuses, and methods for analyzing plant nutrient-yield relationships and identifying optimum plant fertilization strategies to improve both individual and overall plant yields. Unlike previous approaches that only consider nutrient and yield data from the best-performing plants, this invention categorizes and analyzes nutrient and yield data from plants in all yield ranges. The nutrient-yield function for each yield category is modeled using a polynomial equation of various orders, enabling the determination of optimal nutritional conditions for maximizing crop production for each yield category.
The core problem addressed is the lack of effective tools for designing fertilization strategies that improve yields for both high-yielding and low-yielding plants, aiming to enhance overall crop production per unit area or per plant. Traditional models, such as those focusing only on top-producing plants, fail to provide strategies applicable to less productive plants; as a result, overall yield improvement is limited.
The disclosed techniques use quantile regression or artificial neural network (ANN) modeling to generate predictions of potential plant yields in relation to plant nutrient conditions, accounting for plants with inherently different potential yields due to non-nutritional environmental constraints. This approach enables simultaneous optimization of all nutrient elements for plants in different yield categories, providing precise fertilizer recommendations tailored to the specific needs of each category. These strategies can reduce fertilizer costs and improve environmental safety by targeting nutrient applications where they are most effective, thereby minimizing unnecessary fertilization and potential groundwater contamination.
Claims Coverage
The patent presents four main inventive features in its independent claims, focusing on methods, computer-readable media, and systems for determining and applying optimum plant nutritional strategies based on categorized plant data.
Computer-implemented method for improving plant production by categorizing and analyzing nutrient and yield data
The invention implements a method that includes: - Receiving nutrient data (including concentration levels of one or more nutrient elements in plant tissues) and yield data for a plurality of plants. - Categorizing the plants into multiple yield categories based on their yield data. - For each yield category, analyzing the nutrient and yield data to determine a relationship between plant yield and nutrient concentration for each nutrient element. - Applying fertilizers to plants in each yield category with nutrient levels determined by the analyzed relationship for each nutrient element to achieve improved yields.
Non-transitory computer-readable storage medium storing instructions for plant yield optimization based on categorized data analysis
A non-transitory computer-readable storage medium stores instructions that, when executed by one or more processors, cause the system to: - Receive nutrient and yield data for a plurality of plants, including concentration levels of nutrient elements. - Categorize the plants into yield categories. - Analyze, for each yield category, the nutrient and yield data to determine the relationship between plant yield and nutrient concentration for each nutrient element. - Apply fertilizers based on these determined relationships to improve yields for each yield category.
System with processor and instructions for analyzing plant nutrient-yield relationships and applying fertilizers accordingly
A system comprising: - One or more processors and a non-transitory computer-readable medium with instructions to: - Receive nutrient and yield data for a plurality of plants, including nutrient concentration levels. - Categorize plants into yield categories based on yield data. - For each category, analyze nutrient and yield data to determine the relationship between yield and nutrient concentration for each nutrient element. - Apply fertilizers to plants in each yield category using nutrient levels based on these analyzed relationships to achieve improved yields.
The inventive features provide methods, media, and systems for improving plant production by systematically categorizing plants, analyzing nutrient-yield data for each yield category, and applying fertilizers based on precise relationships between yield and nutrient conditions.
Stated Advantages
Allows simultaneous optimization of all nutrient elements for plants of differing potential yields.
Provides targeted fertilization strategies that improve yields for both high-yielding and low-yielding plants, resulting in improved overall crop production.
Reduces fertilizer costs by minimizing unnecessary fertilization.
Improves environmental safety by reducing unnecessary fertilization and minimizing groundwater contamination.
Provides recommendations on fertilization strategies tailored to plant growers for improved yields.
Documented Applications
Improving avocado production through optimized nutrient management.
Enhancing yields in other crops such as pistachios, almonds, citrus, pecans, walnuts, apples, cherries, and peaches.
Interested in licensing this patent?