# US 2022/0036263 A1 · Digital modeling of disease on crops on agronomics fields > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2022/0036263 A1 - Document ID: US20220036263A1 - Status: Published application - Kind: A1 - Application number: 17/502,153 - Filed: 2021-10-15 - Published: 2022-02-03 - Earliest claimed: 2017-11-21 - Assignees: Climate LLC - Applicants: The Climate Corporation - Inventors: Patricia Ann Carroll - Web page: https://patentgrep.com/patent/US20220036263A1 ## CPC - G06Q 10/0635: Risk analysis of enterprise or organisation activities - G06Q 10/067: Enterprise or organisation modelling - G06Q 50/02: Agriculture; Fishing; Forestry; Mining - G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR ## Abstract A system and method for identifying a probability of disease affecting a crop based on data received over a network is described herein, and may be implemented using computers for providing improvements in plant pathology, plant pest control, agriculture, or agricultural management. In an embodiment, a server computer receives environmental risk data, crop data, and crop management data relating to one or more crops on a field. Agricultural intelligence computer system 130 computes one or more crop risk factors based, at least in part, the crop data, one or more environmental risk factors based, at least in part, the environmental data, and one or more crop management risk factors based, at least in part, on the crop management data. Using a digital model of disease probability, agricultural intelligence computer system 130 computes a probability of onset of a particular disease for the one or more crops on the field based, at least in part, on the one or more crop risk factors, the one or more environmental risk factors, and the one or more crop management factors. ## Claims ### Claim 1 A method comprising: receiving environmental risk data relating to a crop on a field; computing one or more environmental risk factors based on the environmental risk data, the one or more environmental risk factors comprising at least one selected from a group consisting of a cumulative disease risk over a period of time and an integral of a cumulative disease risk over a period of time; computing, by a trained model of disease probability, a probability of onset of a particular disease for the crop on the field based on the one or more environmental risk factors, the trained model of disease probability having been trained using a training dataset comprising occurrence or non-occurrence of disease and environmental risk factors comprising the cumulative disease risk and the integral of the cumulative disease risk, wherein the environmental risk factors are used as training inputs and said occurrence or non-occurrence of disease is used as training outputs; based on the on the probability of onset of the particular disease, sending, to an application controller, one or more scripts, wherein the one or more scripts are executed by the application controller to cause an implement on the field to release at least one selected from the group consisting of: water, nutrients, insecticide, and herbicide. ### Claim 2 (depends on 1) The method of claim 1 wherein the environmental risk data indicates a number of risk hours or risk days in which the crop is at risk of developing the particular disease. ### Claim 3 (depends on 2) The method of claim 2 further comprising calculating the cumulative disease risk based on a number of at least one selected from a group consisting of: risk hours, risk days, and growing degree days. ### Claim 4 (depends on 1) The method of claim 1 wherein the trained model of disease probability computing the probability of onset of the particular disease comprises decreasing the probability of onset of the particular disease based on at least one selected from a group consisting of: increased removal of residue by harvesting, increased burial of residue by tilling, and increased crop rotation applied to a portion of the field. ### Claim 5 (depends on 1) The method of claim 1 wherein the trained model of disease probability computing the probability of onset of the particular disease is configured to increase the probability of onset of the particular disease based on at least one selected from a group consisting of: an increased number of days since a last fungicide application on the field, a decreased amount of fungicide applied to the field, and a decreased percentage of the field to which fungicide was applied. ### Claim 6 (depends on 1) The method of claim 1 wherein the training dataset comprises occurrence or non-occurrence of at least one selected from a group consisting of northern leaf blight and gray leaf spot. ### Claim 7 (depends on 1) The method of claim 1 wherein: the crop is based on a type of hybrid seed; the trained model of disease probability computing the probability of onset of the particular disease is based on a value indicating a tolerance of the type of hybrid seed to the particular disease. ### Claim 8 (depends on 1) The method of claim 1 wherein the trained model of disease probability computing the probability of onset of the particular disease is based on information about the field received from at least one selected from a group consisting of: a satellite, a manned aerial vehicle, and an unmanned aerial vehicles. ### Claim 9 (depends on 1) The method of claim 1 further comprising predicting a particular time of onset of the particular disease using the trained model of disease probability. ### Claim 10 (depends on 1) The method of claim 1 wherein the trained model of disease probability computing the probability of onset of the particular disease for the crop on the field comprises a Cox proportional hazard model computing the probability of onset of the particular disease for the crop on the field. ### Claim 11 One or more computer-readable non-transitory media storing instructions that, when executed by one or more processors, cause: receiving environmental risk data relating to a crop on a field; computing one or more environmental risk factors based on the environmental risk data, the one or more environmental risk factors comprising at least one selected from a group consisting of a cumulative disease risk over a period of time and an integral of a cumulative disease risk over a period of time; computing, by a trained model of disease probability, a probability of onset of a particular disease for the crop on the field based on the one or more environmental risk factors, the trained model of disease probability having been trained using a training dataset comprising occurrence or non-occurrence of disease and environmental risk factors comprising the cumulative disease risk and the integral of the cumulative disease risk, wherein the environmental risk factors are used as training inputs and said occurrence or non-occurrence of disease is used as training outputs; based on the on the probability of onset of the particular disease, sending, to an application controller, one or more scripts, wherein the one or more scripts are executed by the application controller to cause an implement on the field to release at least one selected from the group consisting of: water, nutrients, insecticide, and herbicide. ### Claim 12 (depends on 11) The one or more computer-readable non-transitory media of claim 11 wherein the environmental risk data indicates a number of risk hours or risk days in which the crop is at risk of developing the particular disease. ### Claim 13 (depends on 12) The one or more computer-readable non-transitory media of claim 12 wherein the instructions further cause calculating the cumulative disease risk based on a number of at least one selected from a group consisting of: risk hours, risk days, and growing degree days. ### Claim 14 (depends on 11) The one or more computer-readable non-transitory media of claim 11 wherein the trained model of disease probability computing the probability of onset of the particular disease comprises decreasing the probability of onset of the particular disease based on at least one selected from a group consisting of: increased removal of residue by harvesting, increased burial of residue by tilling, and increased crop rotation applied to a portion of the field. ### Claim 15 (depends on 11) The one or more computer-readable non-transitory media of claim 11 wherein the trained model of disease probability computing the probability of onset of the particular disease is configured to increase the probability of onset of the particular disease based on at least one selected from a group consisting of: an increased number of days since a last fungicide application on the field, a decreased amount of fungicide applied to the field, and a decreased percentage of the field to which fungicide was applied. ### Claim 16 (depends on 11) The one or more computer-readable non-transitory media of claim 11 wherein the training dataset comprises occurrence or non-occurrence of at least one selected from a group consisting of northern leaf blight and gray leaf spot. ### Claim 17 (depends on 11) The one or more computer-readable non-transitory media of claim 11 wherein: the crop is based on a type of hybrid seed; the trained model of disease probability computing the probability of onset of the particular disease is based on a value indicating a tolerance of the type of hybrid seed to the particular disease. ### Claim 18 (depends on 11) The one or more computer-readable non-transitory media of claim 11 wherein the trained model of disease probability computing the probability of onset of the particular disease is based on information about the field received from at least one selected from a group consisting of: a satellite, a manned aerial vehicle, and an unmanned aerial vehicles. ### Claim 19 (depends on 11) The one or more computer-readable non-transitory media of claim 11 wherein the instructions further cause predicting a particular time of onset of the particular disease using the trained model of disease probability. ### Claim 20 (depends on 11) The one or more computer-readable non-transitory media of claim 11 wherein the trained model of disease probability computing the probability of onset of the particular disease for the crop on the field comprises a Cox proportional hazard model computing the probability of onset of the particular disease for the crop on the field.