Computer-implemented method for controlling nematode damage

Abstract

Computer-implemented method for controlling nematode damage The present invention describes computer-implemented methods and computer systems where a computer system receives as input field data for an agricultural area, the field data comprising soil data, cultivation data, and nematode characterizing data; where those field data are analyzed by the computer system to determine the risk for nematode damage for each location in the agricultural area; where one or more maps for one or more location in the agricultural area indicating the risk for nematode damage are generated by the computer system; and where one or more maps are provided to a presentation layer.

Description

[0001] COMPUTER-IMPLEMENTED METHOD FOR CONTROLLING NEMATODE DAMAGE

[0002] Background of the invention

[0003] Nematodes are tiny, worm-like, multicellular animals adapted to living in moist surfaces for example in the soil or also within tissues of other organisms or in the water. The number of nematode species is estimated at half a million. As an important part of the soil fauna, nematodes live in a maze of interconnected channels, called pores, that are formed by soil processes. They move in the films of water that cling to soil particles. Plant-parasitic nematodes, a majority of which are root feeders, are found in association with most plants. Some are endoparasitic, living and feeding within the tissue of the roots, tubers, buds, seeds, etc. Others are ectoparasitic, feeding externally through plant walls. A single endoparasitic nematode can kill a plant or reduce its productivity. Endoparasitic root feeders include such economically important pests as the root-knot nematodes (Meloidogyne species), the reniform nematodes (Rotylenchulus species), the cyst nematodes (Heterodera species), and the rootlesion nematodes (Pratylenchus species). Direct feeding by nematodes can drastically decrease a plant’s uptake of nutrients and water. Nematodes have the greatest impact on crop productivity when they attack the roots of seedlings immediately after seed germination. Nematode feeding also creates open wounds that provide entry to a wide variety of plant-pathogenic fungi and bacteria. These microbial infections are often more economically damaging than the direct effects of nematode feeding. Worldwide, losses are estimated to be between 10% and 15%, representing just over $100 billion per year. In Brazil, estimated losses are even greater, exceeding 30%, resulting in over $ 35 billion in losses per year in Soybean, Sugarcane, Cotton, Coffee, Potato, and Carrot crops.

[0004] Detection of nematodes by farmers currently depends on soil or root sampling, which is a time and cost intensive process. In addition, nematode infestation is often very localized in agricultural area thereby increasing the needs for sampling. As nematicide application may be an optional application depending on whether an cultivation conditions and as it represents an additional cost for farmer, there is the need to generate maps showing the risk for nematode damage taking as many as possible of the contributing factors such as the different types of field data into account which allow a targeted application per location using automated prescriptions in an agricultural area to efficiently control nematodes only in those locations where there is a risk for nematode damage and the potential for higher yield. US2019017439, WO2018138648 Al, WO2022 / 223611 also methods for controlling or detecting harmful organism, optionally by predicting infestation risk are described, however nematode control has certain challenges due to the soilborne nature and life cycle of the nematodes. In addition, such a targeted application will also contribute to the objective to reduce the amount of crop protection products in agriculture and may contribute to the overall objective of preserving soil health as well as reducing costs for growers.

[0005] SUMMARY OF INVENTION

[0006] The present invention describes computer-implemented methods and computer systems where a computer system receives as input field data for an agricultural area, the field data comprising soil data, cultivation data, and nematode characterizing data; where those field data are analyzed by the computer system to determine the risk for nematode damage for each location in the agricultural area; where one or more maps for one or more location in the agricultural area indicating the risk for nematode damage are generated by the computer system; and where one or more maps are provided to a presentation layer.

[0007] In one embodiment in an additional step one or more a prescription executable by one or more controlling units of one or more agricultural machinery to apply crop protection solutions suitable for controlling nematodes is generated.

[0008] In one embodiment the field data of the first step comprise nematode susceptibility indices.

[0009] In one embodiment the field data of the first step comprise historic field data.

[0010] In one embodiment the field data are analysed in the second using weighted averages of those field data.

[0011] In one embodiment the risk of nematode damage is used to predict expected yield.

[0012] In one embodiment the risk of nematode damage is used to predict the increase in yield potential.

[0013] In one embodiment the risk for nematode damage or increase in yield potential is used to generate a prescription for a location specific application of a crop protection product.

[0014] DETAILED DESCRIPTION

[0015] The invention will be more described in detail without distinguishing between the aspects of the invention (method, computer system, computer-readable storage medium). On the contrary, the following elucidations are intended to apply analogously to all the aspects of the invention, irrespective of in which context (method, computer system, computer-readable storage medium) they occur.

[0016] Some implementations of the present disclosure are described more fully with reference to the accompanying drawings, in which some, but not all implementations of the disclosure are shown. Various implementations of the disclosure may be embodied in many different forms and should not be construed as limited to those implementations; rather, these example implementations are provided so that this disclosure is thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0017] The operations in accordance with the teachings herein may be performed by at least one computer system specially constructed for the desired purposes or general-purpose computer system specially configured for the desired purpose by at least one computer program stored in a typically non- transitory computer readable storage medium.

[0018] A computer system is a system for electronic data processing that processes data by means of programmable calculation rules. Such a system usually comprises a computer, that unit which comprises a processor for carrying out logical operations, and also peripherals.

[0019] In computer technology, peripherals refer to all devices which are connected to the computer and serve for the control of the computer and / or as input and output devices. Examples thereof are monitor (screen), printer, scanner, mouse, keyboard, drives, camera, microphone, loudspeaker, etc. Internal ports and expansion cards are, too, considered to be peripherals in computer technology.

[0020] Computer systems of today are frequently divided into desktop PCs, portable PCs, laptops, notebooks, netbooks and tablet PCs and so-called handhelds (e. g. smartphone); all these systems can be utilized for carrying out the invention.

[0021] The term non-transitory is used herein to exclude transitory, propagating signals or waves, but to otherwise include any volatile or non-volatile computer memory technology suitable to the application.

[0022] The term computer should be broadly construed to cover any kind of electronic device with data processing capabilities, including, by way of non-limiting example, personal computers, servers, embedded cores, computing system, communication devices, processors (e.g., digital signal processor (DSP)), microcontrollers, field programmable gate array (FPGA), application specific integrated circuit (ASIC), etc.) and other electronic computing devices. The term process as used above is intended to include any type of computation or manipulation or transformation of data represented as physical, e.g., electronic, phenomena which may occur or reside e.g., within registers and / or memories of at least one computer or processor. The term processor includes a single processing unit or a plurality of distributed or remote such units.

[0023] Fig. 1 illustrates a computer system (1) according to some example implementations of the present disclosure in more detail. The computer may include one or more of each of a number of components such as, for example, processing unit (20) connected to a memory (50) (e.g., storage device), communications interface(s) including interface(s) (41) to networks or short-range interfaces configured to connect devices short distance, displays (30), user input interface (11), user interfaces including automatic identification and data capture (AIDC) technology (12). Generally, a computer system of exemplary implementations of the present disclosure may be referred to as a computer and may comprise, include, or be embodied in one or more fixed or portable electronic devices. The computer may include one or more of each of a number of components such as, for example, a processing unit (20) connected to a memory (50) (e.g., storage device).

[0024] The processing unit (20) may be composed of one or more processors alone or in combination with one or more memories. The processing unit (20) is generally any piece of computer hardware that is capable of processing information such as, for example, data, computer programs and / or other suitable electronic information. The processing unit (20) is composed of a collection of electronic circuits some of which may be packaged as an integrated circuit or multiple interconnected integrated circuits (an integrated circuit at times more commonly referred to as a “chip”). The processing unit (20) may be configured to execute computer programs, which may be stored onboard the processing unit (20) or otherwise stored in the memory (50) of the same or another computer.

[0025] The processing unit (20) may be a number of processors, a multi-core processor or some other type of processor, depending on the particular implementation. For example, it may be a central processing unit (CPU), a field programmable gate array (FPGA), a graphics processing unit (GPU) and / or a tensor processing unit (TPU). Further, the processing unit (20) may be implemented using a number of heterogeneous processor systems in which a main processor is present with one or more secondary processors on a single chip. As another illustrative example, the processing unit (20) may be a symmetric multi-processor system containing multiple processors of the same type. In yet another example, the processing unit (20) may be embodied as or otherwise include one or more ASICs, FPGAs or the like. Thus, although the processing unit (20) may be capable of executing a computer program to perform one or more functions, the processing unit (20) of various examples may be capable of performing one or more functions without the aid of a computer program. In either instance, the processing unit (20) may be appropriately programmed to perform functions or operations according to example implementations of the present disclosure.

[0026] The memory (50) is generally any piece of computer hardware that is capable of storing information such as, for example, data, computer programs (e.g., computer-readable program code (60)) and / or other suitable information either on a temporary basis and / or a permanent basis. The memory (50) may include volatile and / or non-volatile memory, and may be fixed or removable. Examples of suitable memory include random access memory (RAM), read-only memory (ROM), a hard drive, a flash memory, a thumb drive, a removable computer diskette, an optical disk, a magnetic tape or some combination of the above. Optical disks may include compact disk - read only memory (CD- ROM), compact disk - read / write (CD-R / W), DVD, Blu-ray disk or the like. In various instances, the memory may be referred to as a computer-readable storage medium or data memory. The computer-readable storage medium is a non-transitory device capable of storing information, and is distinguishable from computer-readable transmission media such as electronic transitory signals capable of carrying information from one location to another. Computer-readable medium as described herein may generally refer to a computer-readable storage medium or computer-readable transmission medium.

[0027] In addition to the memory (50), the processing unit (20) may also be connected to one or more interfaces for displaying, transmitting and / or receiving information. The interfaces may include one or more communications interfaces and / or one or more user interfaces. The communications interface(s) may be configured to transmit and / or receive information, such as to and / or from other computer(s), network(s), database(s) or the like. The communications interface may be configured to transmit and / or receive information by physical (wired) and / or wireless communications links. The communications interface(s) may include interface(s) (41) to connect to a network, such as using technologies such as cellular telephone, Wi-Fi, satellite, cable, digital subscriber line (DSL), fiber optics and the like. In some examples, the communications interface(s) may include one or more short-range communications interfaces (42) configured to connect devices using short-range communications technologies such as NFC, RFID, Bluetooth, Bluetooth LE, ZigBee, infrared (e.g., IrDA) or the like.

[0028] The user interfaces may include a display (30). The display (screen) may be configured to present or otherwise display information to a user, suitable examples of which include a liquid crystal display (LCD), light-emitting diode display (LED), plasma display panel (PDP) or the like. The user input interface(s) (11) may be wired or wireless, and may be configured to receive information from a user into the computer system (1), such as for processing, storage and / or display. Suitable examples of user input interfaces include a microphone, image or video capture device, keyboard or keypad, joystick, touch-sensitive surface (separate from or integrated into a touchscreen) or the like. In some examples, the user interfaces may include automatic identification and data capture (AIDC) technology (12) for machine -readable information. This may include barcode, radio frequency identification (RFID), magnetic stripes, optical character recognition (OCR), integrated circuit card (ICC), and the like. The user interfaces may further include one or more interfaces for communicating with peripherals such as printers and the like.

[0029] As indicated above, program code instructions (60) may be stored in memory (50), and executed by processing unit (20) that is thereby programmed, to implement functions of the systems, subsystems, tools and their respective elements described herein. As will be appreciated, any suitable program code instructions (60) may be loaded onto a computer or other programmable apparatus from a computer-readable storage medium to produce a particular machine, such that the particular machine becomes a means for implementing the functions specified herein. These program code instructions (60) may also be stored in a computer-readable storage medium that can direct a computer, processing unit or other programmable apparatus to function in a particular manner to thereby generate a particular machine or particular article of manufacture. The instructions stored in the computer-readable storage medium may produce an article of manufacture, where the article of manufacture becomes a means for implementing functions described herein. The program code instructions (60) may be retrieved from a computer-readable storage medium and loaded into a computer, processing unit or other programmable apparatus to configure the computer, processing unit or other programmable apparatus to execute operations to be performed on or by the computer, processing unit or other programmable apparatus.

[0030] Retrieval, loading and execution of the program code instructions (60) may be performed sequentially such that one instruction is retrieved, loaded and executed at a time. In some example implementations, retrieval, loading and / or execution may be performed in parallel such that multiple instructions are retrieved, loaded, and / or executed together. Execution of the program code instructions (60) may produce a computer-implemented process such that the instructions executed by the computer, processing circuitry or other programmable apparatus provide operations for implementing functions described herein. Execution of instructions by processing unit, or storage of instructions in a computer -readable storage medium, supports combinations of operations for performing the specified functions. In this manner, a computer system (1) may include processing unit (20) and a computer-readable storage medium or memory (50) coupled to the processing circuitry, where the processing circuitry is configured to execute computer-readable program code instructions (60) stored in the memory (50). It will also be understood that one or more functions, and combinations of functions, may be implemented by special purpose hardware -based computer systems and / or processing circuitry which perform the specified functions, or combinations of special purpose hardware and program code instructions.

[0031] Fig. 2 shows schematically by way of example one embodiment of the computer-implemented method of the present disclosure in the form of a flow chart.

[0032] In the first step (110) of the method (100) the computer system receives field data as input data for an agricultural area, the field data comprising at least soil data, cultivation data, and nematode characterizing data.

[0033] Receives means to accept, retrieve, or obtain one or more of information, data, in particular field data, executable code, or any result from executing an executable code.

[0034] Field data are comprised of the following field data types:

[0035] (a) field identification data such as acreage, field name, field identifiers, geographic identifiers, boundary identifiers, crop identifiers, and any other suitable data that may be used to identify an agricultural area, such as a common land unit (CLU), lot and block number, a parcel number, geographic coordinates and boundaries, Farm Serial Number (FSN), farm number, tract number, field number, section, township, and / or range;

[0036] (b) harvest data such as crop type, crop variety, crop rotation, whether the crop is grown organically, harvest date, Actual Production History (APH), expected yield, yield potential, actual yield, historic yields, crop price, crop revenue, grain moisture, tillage practice, and previous growing season information;

[0037] (c) soil data such as type, composition, pH, organic matter (OM), cation exchange capacity (CEC));

[0038] (d) cultivation data such as planting date, seed(s) type, seed characteristics, resistance profile of crop, relative maturity (RM) of planted seed(s), seed population; (e) treatment application data (e.g., fertilizer data (e.g., nutrient type (Nitrogen, Phosphorous, Potassium), application type, application date, amount, source, method) and / or application of crop protection products (e.g., pesticide, herbicide, fungicide, biologies, other substance or mixture of substances intended for use as a plant regulator, defoliant, or desiccant, application date, amount, source, method, timing),

[0039] (f) irrigation data (e.g., application date, amount, source, method);

[0040] (g) weather data such as precipitation, rainfall rate, predicted rainfall, water runoff rate region, temperature, wind, forecast, pressure, visibility, clouds, heat index, dew point, humidity, snow cover / depth, air quality, sunrise, sunset;

[0041] (h) imagery data such imagery and light spectrum information from an agricultural apparatus sensor, camera, computer, smartphone, tablet, unmanned aerial vehicle, planes or satellite including any data calculated from such imagery such as vegetation indices;

[0042] (i) scouting observations such as photos, videos, free form notes, voice recordings, voice transcriptions, weather conditions like temperature, precipitation (current and over time), soil moisture, crop growth stage, crop physiological stage, phytophysionomy, pest and disease reporting wind velocity, relative humidity, dew point, black layer; and

[0043] (j) nematode characterizing data such as characteristics of a nematode species life cycle, time span of different stages of the nematode, environmental requirements, mobility, nematode population data.

[0044] The term vegetation index means a metric to quantify the health and density of plants or crops calculated on the bases of spectrometric data of the red and near-infrared bands of light. Those data can be sourced from remote sensors e. g. on UAV, in particular drones or satellites. The normalized vegetation index (ND VI) is the quotient of the difference between the reflectance measured at near infrared and red light divided by the sum of these two reflectances.

[0045] The term yield should be construed broadly meaning the amount of harvested plant product such as grain or fruit per unit of agricultural area. In the simplest method the harvested plant material of a defined agricultural area is weighted. In another method yield monitors within agricultural machinery such as combines measure the mass flow of grain e. g. by measuring the weight of grain per time, the volume of grain passing in the combine, or the force of grain hitting a plate in certain time intervals combined with the analysis of positioning while harvesting. Yield may be measured counting or weighing the amount of a harvested crop in one or more sample areas. Grain yield may be measured using a grain yield monitor which is a device coupled to sensors recording the grain or crop yield in the combine or harvester including the measurement of grain flow during the harvesting process including the geolocation.

[0046] Increased yield means an increase in plant product weight, increased plant weight, increased plant height, increased biomass such as higher overall fresh weight (FW), higher grain yield, more tillers, larger leaves, increased shoot growth, or increased content of a specific component present in the plant like increased nutrient content, increased protein content, increased oil content, increased vitamin content, increased starch content, increased pigment content. The amount may be measured in bushel, tons, or kilograms.

[0047] In one embodiment yield includes the yield potential which is the yield of a crop under non-limiting growth conditions including optional water and nutrient availability and effective control of weeds, diseases and pests including nematodes.

[0048] An agricultural area is defined to be any suitable growing space(s) for a crop defined by its boundaries which a) is or is planned to be planted with certain seeds from which the respective crops will grow and harvested or b) is comprise of perennial plants or trees, from which crops such as fruit will be grown and harvested.

[0049] That space may be a greenhouse, an orchard, an experimental plot or any larger area dedicated to grow crops. An agricultural area includes horticultural areas.

[0050] A location is defined to be a subarea of an agricultural area for which the risk assessment is provided. A location may have any size between 10 square meter to 10 ha, preferably 50 square meter to 5 ha, 100 square meter to 1 ha.

[0051] Crops are any plants grown by humans for feed, food or fiber such as cereals like wheat, durum, oats, rye, triticale, barley, rice, corn, oilseeds like soybean, canola, rapeseed, or fiber plants like cotton, hemp or flax, any vegetable or fruit.

[0052] In one embodiment field data comprising at least field identification data, cultivation data, soil data and monitoring data are analyzed using correlation analysis techniques such as univariate analysis, multivariate analysis such as principal component analysis or Pearson’s correlation to identify those field data types which are predictive for the risk of nematode damage and therefore also indicative for yield in any location of an agricultural area. In one embodiment field data comprising at least soil data, cultivation data, and nematode characterizing data are preferably used as input. Those data may include historic data, in particular for cultivation data.

[0053] In another embodiment field data comprising field identification data, cultivation data, soil data and monitoring data are analyzed using machine learning methods to identify those field data types which are predictive for nematode damage and productivity of a field. In one embodiment supervised learning in binary classification models is used. The training data set is comprised field data including soil data, cultivation data and monitoring data.

[0054] In the second step (120) the field data for a certain location are analyzed to determine the risk for nematode damage for each location.

[0055] Nematode damage is the damage on crops including all crop parts like leaves, stems, flowers, seeds, seedlings, runner, buds, roots caused by the presence or activity of one or more nematode or any developmental stage of one or more nematode.

[0056] In the third step one or more maps for one or more location in the agricultural area indicating the risk for nematode damage are generated by the computer system.

[0057] The risk for nematode damage for each location may be determined by analyzing field data.

[0058] In one embodiment the analysis is performed in the disclosed method by using the field data as input data per location to train a model. The output of that model is the risk associated with nematode damage including impact on yield for that location.

[0059] The model may also be or include a machine learning model. A machine learning model as used herein, may be understood as a computer implemented data processing architecture. The machine learning model can receive input data and provide output data based on that input data and on parameters of the machine learning model. The machine learning model can learn a relation between input data and output data through training. In training, parameters of the machine learning model may be adjusted in order to provide a desired output for a given input. In one embodiment the input data are selected from the different types of field data.

[0060] The process of training a machine learning model involves providing a machine learning algorithm (that is the learning algorithm) with training data to learn from. The term trained machine learning model refers to the model artifact that is created by the training process. The training data must contain the correct answer, which is referred to as the target. The learning algorithm finds patterns in the training data that map input data to the target, and it outputs a trained machine learning model that captures these patterns.

[0061] In the training process, training data are inputted into the machine learning model and the machine learning model generates an output. The output is compared with the (known) target. Parameters of the machine learning model are modified in order to reduce the deviations between the output and the (known) target to a (defined) minimum.

[0062] In general, a loss function can be used for training, where the loss function can quantify the deviations between the output and the target. The loss function may be chosen in such a way that it rewards a wanted relation between output and target and / or penalizes an unwanted relation between an output and a target. Such a relation can be, e.g., a similarity, or a dissimilarity, or another relation.

[0063] If, for example, the output and the target are numbers, the loss function could be the difference between these numbers. In this case, a high absolute value of the loss function can mean that a parameter of the model needs to undergo a strong change.

[0064] In the case of vector-valued outputs, for example, difference metrics between vectors such as the root mean square error, a cosine distance, a norm of the difference vector such as a Euclidean distance, a Chebyshev distance, an Lp-norm of a difference vector, a weighted norm or any other type of difference metric of two vectors can be chosen. These two vectors may for example be the desired output (target) and the actual output.

[0065] In the case of higher dimensional outputs, such as two-dimensional, three-dimensional or higherdimensional outputs, for example an element-wise difference metric can be used. Alternatively, or additionally, the output data may be transformed, for example to a one -dimensional vector, before computing a loss.

[0066] The modification of model parameters and the reduction of the loss can be done in an optimization procedure, for example in a gradient descent procedure.

[0067] The model of the present disclosure may by trained on training data. The training data may comprise for each agricultural or horticultural area of a multitude of agricultural or horticultural areas one or more of the different types of field data describing the efficacy of one or more crop protection products on the one or more harmful organisms as target data.

[0068] The term multitude means at least ten, preferably more than a hundred.

[0069] The training of the machine learning model may comprise: inputting the input data into the machine learning model; receiving from the machine learning model a predicted risk for nematode damage; determining a loss, the loss quantifying a deviation between the predicted risk for nematode damage and the target data; modifying parameters of the machine learning model to minimize the loss.

[0070] The machine learning model may be trained on training data for which field data comprising field identification data, cultivation data, soil data and monitoring data are determined in a multitude of agricultural areas at different locations and / or at different time points during the vegetation period and / or over multiple years. Additional data taken into account are the efficacy data of one or more crop protection products of interest at those locations and / or time points.

[0071] The machine learning model may be or comprise an artificial neural network. An “artificial neural network” (ANN) is a biologically inspired computational model. An ANN usually comprises at least three layers of processing elements: a first layer with input neurons (nodes), a kth layer with at least one output neuron (node), and k-2 inner layers, where k is a natural number greater than 2.

[0072] In such a network, the input neurons serve to receive the input data. If the input data constitute or comprise an n-dimensional vector (e.g., a feature vector), with n being an integer equal to or greater than 1 , there is usually one input neuron for each component of the vector. The output neurons serve to output at least one value. The processing elements of the layers are interconnected in a predetermined pattern with predetermined connection weights therebetween. Each network node represents a pre-defined calculation of the weighted sum of inputs from prior nodes and a non-linear output function. The combined calculation of the network nodes relates the inputs to the outputs. It should be noted that an ANN can also use connection biases b, i. e. the output y of a neuron given an input x is calculated as y = act(w x + b) where w denotes the weights and act denotes the activation function.

[0073] When trained, the connection weights between the processing elements in the ANN contain information regarding the relationship between the input data and the output data which can be used to predict new output data from new input data.

[0074] Training estimates network weights that allow the network to calculate output values close to the target values. The network weights can be initialized with small random values or with the weights of a prior partially trained network. The training data inputs are applied to the network and the output values are calculated for each training sample. The network output values are compared to the target values. A backpropagation algorithm can be applied to correct the weight values in directions that reduce the error between targeted and calculated outputs. The process is iterated until no further reduction in error can be made or until a predefined prediction accuracy has been reached.

[0075] A cross-validation method can be employed to split the data into training and validation data sets. The training data set is used in the backpropagation training of the network weights. The validation data set is used to verify that the trained network generalizes to make good predictions. The best network weight set can be taken as the one that best predicts the outputs of the training data. Similarly, varying the number of network hidden nodes and determining the network that performs best with the data sets optimizes the number of hidden nodes.

[0076] Further data can be included in the training of the machine learning model as input data, such as the different types of field data including (a) field identification data such as acreage, field name, field identifiers, geographic identifiers, boundary identifiers, crop identifiers, and any other suitable data that may be used to identify an agricultural area, such as a common land unit (CLU), lot and block number, a parcel number, geographic coordinates and boundaries, Farm Serial Number (FSN), farm number, tract number, field number, section, township, and / or range; (b) harvest data such as crop type, crop variety, crop rotation, whether the crop is grown organically, harvest date, Actual Production History (APH), expected yield, yield potential, actual yield, crop price, crop revenue, grain moisture, tillage practice, and previous growing season information ;(c) soil data such as type, composition, pH, organic matter (OM), cation exchange capacity (CEC));

[0077] (d) cultivation data such as planting date, seed(s) type, seed characteristics, resistance profile of crop, relative maturity (RM) of planted seed(s), seed population; (e) treatment application data (e.g., fertilizer data (e.g., nutrient type (Nitrogen, Phosphorous, Potassium), application type, application date, amount, source, method) and / or application of crop protection products (e.g., pesticide, herbicide, fungicide, biologies, other substance or mixture of substances intended for use as a plant regulator, defoliant, or desiccant, application date, amount, source, method, timing), (f) irrigation data (e.g., application date, amount, source, method); (g) weather data such as precipitation, rainfall rate, predicted rainfall, water runoff rate region, temperature, wind, forecast, pressure, visibility, clouds, heat index, dew point, humidity, snow cover / depth, air quality, sunrise, sunset; (h) imagery data such imagery and light spectrum information from an agricultural apparatus sensor, camera, computer, smartphone, tablet, unmanned aerial vehicle, planes or satellite including any data calculated from such imagery such as vegetation indices; (i) scouting observations such as photos, videos, free form notes, voice recordings, voice transcriptions, weather conditions like temperature, precipitation (current and over time), soil moisture, crop growth stage, crop physiological stage, phytophysionomy, pest and disease reporting wind velocity, relative humidity, dew point, black layer; and (j) nematode characterizing data such as characteristics of a nematode species life cycle, time span of different stages of the nematode, environmental requirements, mobility..

[0078] The term application parameter means any value defining the application of one or more crop protection products including application rate, application method, application timing, application machinery, each for one or more crop protection products.

[0079] When further field data are used, the model learns not only what influence the inclusion of one or more types of field data has on the prediction accuracy for the risk of nematode damage or the increase in yield potential, but also which of these types of field data are most predictive for the accuracy for the risk of nematode damage or the increase in yield potential.

[0080] In one embodiment the model is or comprises a simulation model simulating the impact of one or more type of field data on the risk of nematode damage or the increase in yield potential.

[0081] In one embodiment the risk for nematode damage or the increase in yield potential is determined based on the level of resistance to that crop protection product for a certain variant or a certain genetic profile.

[0082] The expected risk for nematode damage or increase in yield potential can be outputted, e.g., displayed on a display of the computer system, printed via a printing device, stored in a data memory and / or transmitted to a separate computer system.

[0083] Based on the outputted expected efficacy, a user (e.g., a farmer) can see whether the crop protection product for which the expected efficacy has been outputted is a suitable product to control the harmful organisms in the agricultural or horticultural area or whether the efficacy is too low, e. g. because harmful organisms are present in the agricultural or horticultural area that have developed resistance to the crop protection product.

[0084] The user can display expected efficacies of a plurality of crop protection products to determine which crop protection product has the highest efficacy.

[0085] The computer system / computer program of the present disclosure may also be configured to determine the efficacy of a plurality of crop protection products and identify a number of crop protection products that have the highest efficacies, e.g., a number of 1, 2, 3, 4, or 5 or more crop protection products.

[0086] The computer system / computer program of the present disclosure may be configured to output the number of pesticides with the highest efficacies. In addition to one or more crop protection products, a quantity or quantities may be outputted that should be applied to effectively control harmful organisms and prevent resistance.

[0087] If different harmful organisms or different variants of one or more harmful organisms have been identified at different locations in the agricultural or horticultural area, the computer system / computer program may be configured to output, for each location, one or more crop protection products that are most effective in controlling the harmful organisms or variants at the location.

[0088] For example, a map of the agricultural or horticultural area can be output showing, for different subareas of the agricultural or horticultural area, which crop protection products have the highest efficacies against harmful organisms or variants in the subarea.

[0089] The user may be prompted to select a crop protection product from a list of crop protection products for each subarea. However, it is also possible that the computer system / computer program is configured to select only one crop protection product for each subarea (e.g., the one with the highest efficacy) and list it on a map of the agricultural or horticultural area.

[0090] The computer system / computer program may also be configured to specify, for subareas, the amounts of a crop protection product in which it should be applied to the subarea to control the harmful organisms present there.

[0091] The computer system / computer program may also be configured to output an application map, i.e., a map of the agricultural or horticultural area in which for subareas the amount of one or more crop protection products is specified which are to be applied in the subareas to control one or more harmful organisms and / or to prevent resistance.

[0092] Such an application map can also be transmitted over a network to an agricultural or horticultural machinery such as a vehicle, drone, or robot that performs the application of the one or more crop protection products according to the application map. Such an application map is also referred to as prescription.

[0093] A prescription means a script capable of executing commands on one or more type of agricultural machinery so that the agricultural machinery performs its function both in time and location, optionally by automatic steering using a GPS signal to steer the agricultural machinery. In one embodiment the agricultural machinery is a spraying device. Examples for spraying devices are self- propelled sprayers, tractor mounted spraying devices, backpack spraying devices, ATV spraying devices. In one embodiment a prescription is used to control a spraying device suitable to spray one or more crop protection products to spray certain amounts of one or more crop protection products at a certain time and / or a certain location with certain application parameters. Those application parameters include the type of spraying device recommended, additional components for the spraying device such as boom type, nozzles, the spraying timing and location direction, the speed of the travelling spraying device, the number of passes for an application, the application rate, or the cell size. The spraying devices may be equipped with selective or automatic section control allowing to avoid double spray over overlap eg at field ends. They may also be equipped with rate control systems such as pulse with modulation allowing spraying a uniform volume. The cell size is an important parameter in a prescription, it is the size of the cell of the grid which is layered on top of a resistance or genetic profile map to determine whether a crop protection product should be applied in that cell (JC Mayer “USING PRESCRIPTION MAPS FOR IN FIELD EVALUATION OF PARAMETERS AFFECTING SPRAYING ACCURACY OF A SELF A PROPELLED SPRAYER“, Master Thesis North Dacota State University of Agriculture and Applied Science, 2021). A prescription may be machine readable or may be read by humans as well. A prescription may visualized as a prescription map to a user.

[0094] In one embodiment it is determined using a multicriteria analysis methodology in which variables are combined through a weighted average. In this methodology, the weights for each type of field data may be determined by different methods such a Fuzzy logic in order to select the best type of field data predicting nematode damage and to determine the optimal weighing of those field data. In one embodiment at least 1 to 10 type of field data are used comprising soil data, imagery data from which further parameter such vegetation indices generated from imagery data may be used for the analysis.

[0095] In one embodiment the risk for nematode damage is defined in categories such as no risk, medium risk, high risk.

[0096] In the third step (130) one or more maps disclosing indicating the risk for nematode damage per location are generated by the computer system.

[0097] In the fourth step (140) one or more maps disclosing indicating the risk for nematode damage per location are provided by the computer system to a presentation layer.

[0098] In one embodiment, the computer system is programmed with or comprises a communication layer, presentation layer, data management layer, hardware / virtualization layer 150, and model and field data repository. Layer means to any combination of electronic digital interface circuits, microcontrollers, firmware such as drivers, and / or computer programs or other software elements.

[0099] The communication layer may be programmed or configured to perform input / output interfacing functions including external data server computer, or remote sensor for field data. Communication layer may be programmed or configured to send the received data to model and field data repository 160 to be stored as field data 106.

[0100] The presentation layer may be programmed or configured to generate a graphical user interface (GUI) to be displayed on the same or different computer system including a cab computer or a mobile that are coupled to the computer system directly or remotely. The GUI may comprise controls for providing field data to be sent to the computer system, generating, providing or displaying one or more maps or one or more prescriptions.

[0101] In one embodiment one or more maps are displayed on the presentation layer. In another embodiment a visualization of one or more prescriptions is displayed on the presentation layer.

[0102] In one embodiment in an additional step one or more a prescription executable by one or more controlling units of one or more agricultural machinery to apply crop protection solutions suitable for controlling nematodes is generated using one or more maps.

[0103] A crop protection solution means any combination of one or more crop protection products in particular one or more nematicides with (i) instructions regarding timing and application for those nematicides optionally in combination with other crop protection products or (ii) with seeds for suitable varieties or hybrids.

[0104] In one embodiment the crop protection products comprise one or more chemical or biological nematicides selected from the group comprising abamectin, Bacillus subtilis, Bacillus subtilis strain 2657 (Furatrop™) fluensulfone, fluopyram, cyclobutrifluram, Bacillus firmus, Bacillus firmus, imicyafos, fosthiazate, fluazaindolizine.

[0105] In one embodiment a prescription generated by the computer system provides parameters like the application rate for the application of a nematicide to a sprayer, preferably for the application of at least one crop protection product comprising abamectin, fluensulfone, fluopyram, cyclobutrifluram, Bacillus firmus, Bacillus firmus, imicyafos, fosthiazate, fluazaindolizine. In one embodiment a prescription generated by the computer system provides parameters like the application parameters for the application of a nematicide to a sprayer, preferably for the application of at least one crop protection product comprising fluopyram, cyclobutrifluram, fluazaindolizine.

[0106] In one embodiment a prescription generated by the computer system provides parameters like the application rate for the application of a nematicide to a sprayer, preferably for the application of at least one crop protection product comprising fluopyram.

[0107] In one embodiment one or more prescriptions executable by one or more controlling units of one or more agricultural machinery to apply one or more nematicides suitable for controlling nematodes is generated using one or more maps.

[0108] In one embodiment one or more prescriptions executable by one or more controlling units of one or more agricultural machinery to (i) in a first step plant a certain crop variety and (ii) in a second step apply one or more nematicides suitable for controlling nematodes is generated using one or more maps.

[0109] Nematicides are either chemical or biological active ingredient suitable of controlling phytopathogenic nematodes. Examples for Nematicides are Fluopyram (Brand Name Verango or Velum), abamectin, fluensulfone, fluopyram, cyclobutrifluram, Bacillus firmus, Bacillus firmus, imicyafos, fosthiazate, fluazaindolizine.

[0110] Agricultural machinery means apparatuses either autonomous or controlled by human such as tractors, combines, harvesters, planters, trucks, fertilizer equipment, aerial vehicles including unmanned aerial vehicles, and any other item of physical machinery or hardware, typically mobile machinery, and which may be used in tasks associated with agriculture.

[0111] In one embodiment the field data used in the first step comprise nematode susceptibility indices.

[0112] In one embodiment a Nematode susceptibility index is calculated on the basis of a specific vegetation index, the soil composition, and the water height above minimum drainage.

[0113] In another embodiment the species specific nematode susceptibility index is calculated for a species selected from Pratylenchus, Meloidogyne,

[0114] In one embodiment the field data of the first step comprise historic field data.

[0115] In one embodiment the field data of the first step comprise historic or current field data.

[0116] In one embodiment the field data of the first step comprise historic or current field data or both. Historic field data means field data which has been generated in previous growing seasons.

[0117] Current field data means field data which are or have been generated within this growing season starting from the preparation of the agricultural area for planting of the crops until the harvest.

[0118] In one embodiment the field data are analysed using weighted averages of those field data.

[0119] In one embodiment the expected yield is predicted using the risk of nematode damage. Several methods including machine learning methods as described above may be used to predict the expected yield from the risk of nematode damage. Examples are statistical method such as bivariate analysis including correlation analysis, regression analysis, chi-square tests or t-tests or ANOVA or multivariate analysis including Principal Component Analysis, Multivariate analysis of variance (MANOVA), cluster analysis, logistic or multiple linear regression. The variable for those methods is the risk of nematode damage and one or more variables selected from the field data, in particular soil variables such as type, composition, pH, organic matter (OM), cation exchange capacity (CEC)); nematode characterizing data such as characteristics of a nematode species life cycle, time span of different stages of the nematode, environmental requirements, mobility, nematode population data; harvest data such as crop type, crop variety, crop rotation, whether the crop is grown organically, harvest date, Actual Production History (APH), yield potential, actual yield, historic yields, crop price, crop revenue, grain moisture, tillage practice, and previous growing season information.

[0120] In one embodiment the increase in yield potential is predicted using the risk of nematode damage.

[0121] Several methods including machine learning methods as described above may be used to predict the increase in yield potential from the risk of nematode damage. Examples are statistical method such as bivariate analysis including correlation analysis, regression analysis, chi-square tests or t-tests or ANOVA or multivariate analysis including Principal Component Analysis, Multivariate analysis of variance (MANOVA), cluster analysis, logistic or multiple linear regression. The variable for those methods is the risk of nematode damage and one or more variables selected from the field data, in particular soil variables such as type, composition, pH, organic matter (OM), cation exchange capacity (CEC)); nematode characterizing data such as characteristics of a nematode species life cycle, time span of different stages of the nematode, environmental requirements, mobility, nematode population data; harvest data such as crop type, crop variety, crop rotation, whether the crop is grown organically, harvest date, Actual Production History (APH), expected yield, actual yield, historic yields, crop price, crop revenue, grain moisture, tillage practice, and previous growing season information. In the third step one or more maps indicating the nematode damage at per location for one or more agricultural areas are generated, which may be provided to the presentation layer.

[0122] In the last step a prescription executable by one or more controlling units of one or more agricultural machinery in order to apply crop protection products is generated, in one embodiment using a computer system.

[0123] EXAMPLES

[0124] Example A

[0125] The model was built and validated using data from 15.000 ha spread around two major soy producer macro regions in Brazil and currently suffer substantial losses caused by nematode. Analyzing the harvest data from fields spread in those regions and correlating it to nematode damage maps based on the location and analysis of soil samples and the model prescription map, it was compared (overlapped) all 3 maps (nematode damage zone, model prescription and yield. Within each nematode damage zone (high, medium or low) it was compared grid cells indicated by the model to apply or not apply and compared its yield with its adjacent area not applied. If the model indicated to apply in the in the analyzed area and the yield difference was higher than than a certain threshold compared to the yield of its adjacent control area that wasn't sprayed, would count as a win for the model (see picture below). If the difference was lower than 1 bag it was considered a loss for the model. The final model accuracy was satisfactory enough to move it for a pilot test.

[0126] Fig 4 A describes an example for an application map disclosing in which areas a nematicide application is not recommended or recommended using the standard application parameter for that product, here Verango.

[0127] Brown location: area not recommended for a nematicide application, in this case the product Verango comprising Fluopyram, using the prediction of the model;

[0128] Green location: area indicated to be sprayed using the prediction of the model;

[0129] Red strip of locations: agricultural area used as a control which means no nematicide was applied;

[0130] Blue strip of locations: agricultural area used for the model evaluation.

[0131] Example B

[0132] A soybean plot which has been subject to nematode damage in the past was planted with the commercial soy variety Monsoy 6100 XTD from Sementes Copercampos, Brasil during the summer season 2022 / 2023 on 30 October 2022 in the Mato Grosso region of Brazil. The boundaries of the plot location was obtained using the shapefile format. The Monsoy 6100 XTD variety has a G.M (maturity grade) of 6.1 and is susceptible to nematode damage. The same location was planted with soy and treated with biological products for nematode control using a spray application in the previous season. Within 72 hours and using all the information above provided by the farmer and the plot 's environmental and agronomical specs obtained by the model, A prescription for the application of the nematicide Verango (active ingredient Fluopyram) was generated for that location using 1 hectar grid and indicating specific zones within the plot with higher potential for yield increase if sprayed with it.

[0133] Fig 4 B describes an example for an application map disclosing in which areas a nematicide application is not recommended or recommended using the standard application parameter for that product, here Verango.

[0134] Brown location: agricultural area with lower increase of yield potential and not recommended for Verango application based the model;

[0135] Green location: agricultural area with higher yield increase potential and indicated to be sprayed based the model;

[0136] Then, the farmer would do a zonal spray following Verango recommendation. A boom spray would be done after planting and before plant emergence, via zonal application. The farmer needs to have the section spray enabled in his machinery.

[0137] If desired (optional), check strips could be used to evaluate the ROI (return over investment) by comparing the average yield of the sprayed control area (not sprayed with Verango) with its adjacent sprayed area.

[0138] By the season's end, using FieldView (Bayer tool for farming management) and calibrated harvesters, the grower would analyze the collected yield data to verify the gains over the sprayed zones. In this case, the average yield increase checked was 5 bags per hectare, which not only paid for the application cost but provided extra income for the farmer and proved the benefits of zonal application. Example 3

[0139] Fig. 3 shows schematically by way of example the process of training a machine learning model.

[0140] The machine learning model MLM is trained on the basis of training data TD.

[0141] The training data TD comprise for each agricultural area of a multitude of agricultural areas i) one or more types of field data FD and ii) nematode risk data. The training data TD may comprise further data FD as described herein.

[0142] The one or more genetic profiles GP and optionally the further data FD are inputted into the machine learning model MLM. The machine learning model is configured to generate, at least partially on the basis of the inputted data and model parameters, the predicted risk for nematode damage data pND or predicted increase in yield potential for one or more location. The predicted risk for nematode damage data pND or predicted increase in yield potential are compared with the field data comprising nematode damage ND or yield potential YP. This is done by using a loss function LF, the loss function quantifying the deviations between the predicted risk for nematode damage data pND or predicted increase in yield potential and the nematode damage ND and yield potential YP. For each pair of the predicted risk for nematode damage data pND or predicted increase in yield potential and the nematode damage ND and yield potential YP, a loss value is computed. During training the model parameters are modified in a way that reduces the loss values to a defined minimum. The aim of the training is to let the machine learning model generate for each input data an output which comes as close to the corresponding target as possible. Once the defined minimum is reached, the (now fully trained) machine learning model can be used to predict an output for new input data (input data which have not been used during training and for which the target is usually not (yet) known).

Claims

SET OF CLAIMS1. Computer- implemented method comprising• Receiving at a computer system field data as input data for an agricultural area, the field data comprising soil data, cultivation data, and nematode characterizing data;• Analyzing by the computer system those field data to determine the risk for nematode damage for each location in an agricultural area;• Generating by the computer system one or more maps for one or more locations in the agricultural area indicating the risk for nematode damage per location;• Providing one or more maps to a presentation layer;2. The method according to claim 1, wherein in an additional step one or more prescription executable by one or more controlling units of one or more agricultural machinery to apply crop protection solutions suitable for controlling nematodes are generated.

3. The method according to claim 1 or 2, wherein in the first step the field data comprise nematode susceptibility indices.

4. The method according to claims 1 to 3 where in in the first step the field data comprise historic or current field data or both.

5. The method according to any of claims 1 to 4 wherein the field data are analyzed using weighted averages of those field data.

6. The method according to any of claims 1 to 5 wherein the risk of nematode damage is used to predict expected yield.

7. The method according to any of claims 1 to 5 wherein the risk of nematode damage is used to predict the increase in yield potential.

8. Computer system comprising at least one processor programmed to• Receive as input field data for an agricultural area, the field data comprising soil data, cultivation data, and nematode characterizing data;• Analyze those field data to determine the risk for nematode damage for each location in an agricultural area;• Generating one or more maps for one or more location in the agricultural area indicating risk for nematode damage per location;• Provide one or more maps to a presentation layer9. The computer system according to claim 7 comprising at least one processor programmed to generate one or more prescriptions executable by one or more controlling units of one or more agricultural machinery to apply crop protection products suitable for controlling nematodes.

10. The computer system according to claim 7 or 8 wherein the field data used as input comprise nematode susceptibility indices.

11. The computer system according to any of claim 7 to 9 wherein the field data comprise historic or current data or both.

12. The computer system according to any of claims 7 to 10 wherein the field data are analyzed using weighted averages of those field data.

13. The computer system according to any of claims 7 to 11 wherein the risk of nematode damage is used to predict expected yield.

14. The computer system according to any of claims 7 to 11 wherein the risk of nematode damage is used to predict the increase in yield potential.

Images