Search

EP-4075939-B1 - SYSTEMS AND METHODS FOR MODELING DISEASE SEVERITY

EP4075939B1EP 4075939 B1EP4075939 B1EP 4075939B1EP-4075939-B1

Inventors

  • RUSSO, Joseph, Martin

Dates

Publication Date
20260506
Application Date
20201218

Claims (15)

  1. A non-transitory computer-readable medium storing instructions that, when executed by one or more processors (702), cause the one or more processors to: receive, at a computing device, an input (200) comprising weather data (202) for a given day; conduct, using the computing device, a simulation using a disease severity model (300) at predefined intervals for the given day, the simulation adjusting a canopy moisture budget for a plant (102) at each of the predefined intervals; determine, using the computing device, a leaf wetness duration based on the canopy moisture budget, the leaf wetness duration representing a period of time during which leaves of the plant (102) are assumed to be wet; determine, using the computing device, an average air temperature based on the leaf wetness duration; compute, using the computing device, a disease severity level based on the leaf wetness duration and the average air temperature; and generate, at the computing device, an output (314) comprising the disease severity level for the given day.
  2. The medium of claim 1, further storing instructions for determining one or more of a precipitation amount, a dew amount, or a leaf surface evaporation amount, wherein precipitation and dew add to the canopy moisture budget and leaf surface evaporation subtracts from it.
  3. The medium of claim 2, wherein: the weather data (202) describes the precipitation amount, a relative humidity (210), a wind speed (212), and an amount of incoming solar radiation (214); the dew amount is increased when the relative humidity (210) exceeds a predetermined threshold; and the leaf surface evaporation is a function of the relative humidity (210), the wind speed (212), and the incoming solar radiation (214).
  4. The medium of claim 1, wherein the disease severity level is determined to increase when the average air temperature falls within a predetermined range during the leaf wetness duration.
  5. The medium of claim 1, wherein computing the disease severity level comprises determining a minimum number of wetness hours (312) for infection, determining whether the leaf wetness duration exceeds the minimum number of wetness hours (312), and determining the disease severity level based on the leaf wetness duration exceeding the minimum number of wetness hours at a favorable air temperature.
  6. The medium of claim 1, wherein the disease severity level is defined by a length of the leaf wetness duration.
  7. The medium of claim 1, wherein: a single disease event occurs in the given day, and the disease severity level is set to a severity of the single disease event; or a plurality of disease events having different severities occur in a given day, and the disease severity level is set to a severity of a most-severe event.
  8. A method comprising: receiving, at a computing device, an input (200) comprising weather data (202) for a given day; conducting, using the computing device, a simulation using a disease severity model (300) at predefined intervals for the given day, the simulation adjusting a canopy moisture budget for a plant (102) at each of the predefined intervals; determining, using the computing device, a leaf wetness duration based on the canopy moisture budget, the leaf wetness duration representing a period of time during which leaves of the plant (102) are assumed to be wet; determining, using the computing device, an average air temperature based on the leaf wetness duration; computing, using the computing device, a disease severity level based on the leaf wetness duration and the average air temperature; and generating, at the computing device, an output (314) comprising the disease severity level for the given day.
  9. The method of claim 8, further storing instructions for determining one or more of a precipitation amount, a dew amount, or a leaf surface evaporation amount, wherein precipitation and dew add to the canopy moisture budget and leaf surface evaporation subtracts from it.
  10. The method of claim 9, wherein: the weather data (202) describes the precipitation amount, a relative humidity (210), a wind speed (212), and an amount of incoming solar radiation (214); the dew amount is increased when the relative humidity (210) exceeds a predetermined threshold; and the leaf surface evaporation is a function of the relative humidity (210), the wind speed (212), and the incoming solar radiation (214).
  11. The method of claim 8, wherein the disease severity level is determined to increase when the average air temperature falls within a predetermined range during the leaf wetness duration.
  12. The method of claim 8, wherein computing the disease severity level comprises determining a minimum number of wetness hours (312) for infection, determining whether the leaf wetness duration exceeds the minimum number of wetness hours (312), and determining the disease severity level based on the leaf wetness duration exceeding the minimum number of wetness hours at a favorable air temperature.
  13. The method of claim 8, wherein the disease severity level is defined by a length of the leaf wetness duration.
  14. The method of claim 8, wherein: a single disease event occurs in the given day, and the disease severity level is set to a severity of the single disease event; or a plurality of disease events having different severities occur in a given day, and the disease severity level is set to a severity of a most-severe event.
  15. An apparatus comprising: a hardware interface configured to receive an input (200) comprising weather data (202) for a given day; a non-transitory computer readable medium storing a disease severity model (300) configured to be simulated at predefined intervals for the given day, the simulation adjusting a canopy moisture budget for a plant (102) at each of the predefined intervals; a hardware processor (702) configured to; determine a leaf wetness duration based on the canopy moisture budget, the leaf wetness duration representing a period of time during which leaves of the plant (102) are assumed to be wet; determine an average air temperature based on the leaf wetness duration; compute a disease severity level based on the leaf wetness duration and the average air temperature; and generate an output (314) comprising the disease severity level for the given day.

Description

CROSS REFERENCE TO RELATED APPLICATION(S) This application claims the benefit of priority of U.S. Non-Provisional Patent Application No. 16/720,713, filed on December 19, 2019. FIELD OF THE INVENTION The present application relates to computational techniques for predicting, detecting, and modeling diseases in plants and animals. BACKGROUND US 2018/0262571 A1 describes an integrated IoT (Internet of Things) system for Smart Agriculture management increasing crop yield, optimize food storage, distribution and delivery using IoT and Artificial Intelligence in its communication and supply chain infrastructure. STREIT KATARINA ET AL, "Modelling the combined effect of moisture and temperature on secondary infection in a coupled host-pathogen FSPM", 2018 6TH INTERNATIONAL SYMPOSIUM ON PLANT GROWTH MODELING, SIMULATION, VISUALIZATION AND APPLICATIONS (PMA) describes simulating the effect of weather conditions on the progression of secondary infection in yellow spot and the interaction with growing wheat canopy, XP033501623. SIMONE BREGAGLIO ET AL, "Multi metric evaluation of leaf wetness models for large-area application of plant disease models", AGRICULTURAL AND FOREST METEOROLOGY, ELSEVIER, AMSTERDAM, NL, vol. 151, no. 9, describes evaluating models for the estimation of leaf wetness (LW) in large-area scenario analysis. Within this framework, the following specific objectives were carried out: (i) comparison of six LW models; and (ii) assessment of the impact of LW estimated data as input on an impact model, XP028233423. JACOBS A F GET AL, "Simulating of leaf wetness duration within a potato canopy", NJAS - WAGENINGEN JOURNAL OF LIFE SCIENCES, ELSEVIER, AMSTERDAM, NL, vol. 53, no. 2 describes applying within-canopy dew simulation model to simulate leaf wetness distribution in the canopy caused by dew and rainfall, XP026644771. In agriculture it is desirable to maximize, as much as possible, the production or yield of a given agricultural product. One factor that limits the amount of production or yield are diseases, which can include (for example) micro-organisms, insects, bacteria, fungi, viruses, infectious agents, parasites, and genetic disorders. A disease outbreak can reduce or even outright destroy a harvest. In many cases, agricultural diseases can be treated or controlled - for example, by spraying a crop with an appropriate treatment such as a fungicide, insecticide, or pesticide. However, these treatments are expensive and time-consuming to apply. Accordingly, they should be applied only where and when they can be most effective. In order to effectively target disease treatments, it is critical to understand (and, if possible, to predict) the risk of a disease outbreak at a given time and place. A number of techniques exist for simulating a disease outbreak. Typically, these simulations attempt to model an exact relationship between a particular type of disease and a particular host. Accordingly, different simulations are needed to model the various diseases that could affect a single type of crop; if multiple crop types must be considered, the number of simulations increases proportionally. Furthermore, each individual simulation tends to be computationally intensive, because a large number of variables and parameters must be accounted for in order to model an exact relationship between a disease and a host. Thus, existing simulation methods require a significant amount of time in order to evaluate the risk to a given agricultural product, which reduces the time available to respond to a predicted outbreak. Moreover, a great deal of research and work needs to be done to determine the relationship between a given disease and a particular host. This makes it difficult to expand a simulation to new disease types. SUMMARY According to a first embodiment, a method for predicting, detecting, and modeling diseases in plants and animals is provided. A computing device may receive an input comprising weather data for a given day. The computing device may conduct a simulation using a disease severity model at predefined time intervals for a given day that uses air temperature and leaf wetness duration as input. The leaf wetness duration is derived from a canopy moisture budget and is defined as a continuous period when a leaf canopy is wet. The air temperature may be averaged during a leaf wetness duration at predefined time intervals during that period. The computing device may compute a disease severity level based on the leaf wetness duration and the average air temperature. The computing device may generate an output comprising the disease severity level for the given day. A second embodiment incorporates the method of the first embodiment, and further comprises determining one or more of a precipitation amount, a dew amount, or a leaf surface evaporation amount, wherein precipitation and dew add to the canopy moisture budget and leaf surface evaporation subtracts from it. A third embodiment incorporates t