Fractional ice cover predictions with machine learning, satellite, thermodynamics, and in-situ observations

{
  "publication_number": "US12429628B2",
  "country": "US",
  "kind": "B2",
  "title": "Fractional ice cover predictions with machine learning, satellite, thermodynamics, and in-situ observations",
  "abstract": "A computer implemented method of predicting ice coverage on a body of water includes generating a first ice cover prediction with a thermodynamics module and generating a second ice cover prediction with a machine learning module. The first ice cover prediction is combined with the second ice cover prediction to generate a combined ice cover prediction. Error statistics are computed based on a comparison of the combined ice cover prediction with an ice coverage observation and the combined ice cover prediction is updated based on the error statistics.",
  "claims": [
    "1. A computer implemented method of predicting ice coverage on a body of water, comprising: generating a first ice cover prediction on a first map grid, with respect to the body of water, with a thermodynamics module; generating a second ice cover prediction on a second map grid, with respect to the body of water, with a machine learning module; combining the first ice cover prediction on the first map grid and the second ice cover prediction on the second map grid to generate a combined ice cover prediction on a combined map grid, with respect to the body of water; computing error statistics based on a comparison of the combined ice cover prediction with an ice coverage observation; updating the combined ice cover prediction on the combined map grid based on the error statistics; and updating a travel path of an automated observation device based on the combined ice cover prediction on the combined map grid.",
    "2. The computer implemented method of claim 1, wherein the thermodynamics module includes a hydrodynamic model and one or more thermodynamic equations used to generate the first ice cover prediction.",
    "3. The computer implemented method of claim 2, further comprising updating the hydrodynamic model based on the error statistics.",
    "4. The computer implemented method of claim 1, wherein the machine learning module includes a trained model and a module for generating a predicted local weather condition.",
    "5. The computer implemented method of claim 4, further comprising providing a feedback loop to the trained model based on the error statistics.",
    "6. The computer implemented method of claim 1, further comprising applying weightings and thresholds to generate the combined ice coverage prediction from the first ice coverage prediction and the second ice coverage prediction.",
    "7. The computer implemented method of claim 6, wherein the weightings and thresholds are applied to each grid cell defining a surface area of the body of water.",
    "8. The computer implemented method of claim 6, wherein the weighting and thresholds for each grid cell are defined separately based on a predefined period throughout a year.",
    "9. The computer implemented method of claim 6, wherein a tree-based model is used to optimize the combined ice coverage prediction.",
    "10. The computer implemented method of claim 1, further comprising performing direct observations of ice coverage at the body of water with the observation device.",
    "11. The computer implemented method of claim 10, wherein a travel path for the observation device is optimized based on the combined ice coverage prediction and observations of whether ice is present at a grid cell defining a surface area of the body of water.",
    "12. The computer implemented method of claim 11, further comprising updating the travel path when disagreement occurs regarding ice coverage based on the combined ice coverage prediction and the observations of the observation device.",
    "13. The computer implemented method of claim 10, wherein the observation device is an in-situ automated observation device.",
    "14. A computer implemented method for predicting ice coverage on a body of water, comprising: generating a first ice cover prediction on a first map grid, with respect to the body of water, with a thermodynamics module; generating a second ice cover prediction on a second map grid, with respect to the body of water, with a machine learning module; combining the first ice cover prediction on the first map grid and the second ice cover prediction on the second map grid to generate a combined ice cover prediction on the combined map grid based on an initial set of weightings and thresholds; computing error statistics based on a comparison of the combined ice cover prediction with an ice coverage observation from an automated observation device; updating the weightings and thresholds based on the error statistics; using the updated weightings and thresholds to generate an updated combined ice cover prediction on the combined map grid; and updating a travel path of an automated observation device based on the combined ice cover prediction on the combined map grid.",
    "15. The computer implemented method of claim 14, further comprising updating a travel path of the automated observation device when disagreement occurs regarding ice coverage based on the combined ice coverage prediction and the observations made of ice coverage with the automated observation device.",
    "16. The computer implemented method of claim 14, wherein the weightings and thresholds are applied to each grid cell defining a surface area of the body of water.",
    "17. The computer implemented method of claim 14, wherein the weighting and thresholds for each grid cell are defined separately based on a predefined period throughout a year.",
    "18. The computer implemented method of claim 14, wherein a tree-based model is used to optimize the combined ice coverage prediction.",
    "19. A non-transitory computer readable storage medium tangibly embodying a computer readable program code having computer readable instructions that, when executed, causes a computer device to carry out a method of predicting ice coverage on a body of water, the method comprising: generating a first ice cover prediction on a first map grid, with respect to the body of water, with a thermodynamics module; generating a second ice cover prediction on a second map grid, with respect to the body of water, with a machine learning module; combining the first ice cover prediction on the first map grid and the second ice cover prediction on the second map grid to generate a combined ice cover prediction on a combined map grid, with respect to the body of water; computing error statistics based on a comparison of the combined ice cover prediction with an ice coverage observation; updating the combined ice cover prediction on the combined map grid based on the error statistics; and updating a travel path of an automated observation device based on the combined ice cover prediction on the combined map grid.",
    "20. The non-transitory computer readable storage medium of claim 19, wherein the execution of the code by the processor further configures the computing device to perform an act comprising: generating the combined ice coverage prediction by applying weightings and thresholds to the first ice coverage prediction and the second ice coverage prediction.",
    "21. The non-transitory computer readable storage medium of claim 19, wherein the execution of the code by the processor further configures the computing device to perform an act comprising: performing direct observations of ice coverage at the body of water, wherein automated observation device is an in-situ automated observation device.",
    "22. A non-transitory computer readable storage medium tangibly embodying a computer readable program code having computer readable instructions that, when executed, causes a computer device to carry out a method of predicting ice coverage on a body of water, the method comprising: generating a first ice cover prediction on a first map grid, with respect to the body of water, with a thermodynamics module; generating a second ice cover prediction on a second map grid, with respect to the body of water, with a machine learning module; combining the first ice cover prediction on the first map grid and the second ice cover prediction on the second map grid to generate a combined ice cover prediction on a combined map grid, with respect to the body of water, based on an initial set of weightings and thresholds; computing error statistics based on a comparison of the combined ice cover prediction with an ice coverage observation from an automated observation device; updating the weightings and thresholds based on the error statistics; using the updated weightings and thresholds to generate an updated combined ice cover prediction on the combined map grid; and updating a travel path of the automated observation device based on the combined ice cover prediction on the combined map grid.",
    "23. The non-transitory computer readable storage medium of claim 22, wherein the execution of the code by the processor further configures the computing device to perform an act comprising: updating a travel path of the automated observation device when disagreement occurs regarding ice coverage based on the combined ice coverage prediction and the observations made of ice coverage with the automated observation device.",
    "24. The non-transitory computer readable storage medium of claim 22, wherein the weightings and thresholds are applied to each grid cell defining a surface area of the body of water.",
    "25. The non-transitory computer readable storage medium of claim 24, wherein the weighting and thresholds for each grid cell are defined separately based on a predefined period throughout a year."
  ],
  "description_excerpt": "The present disclosure generally relates to artificial intelligence and machine learning systems, and more particularly, to methods and systems for predicting fractional ice cover on a body of water.\n\nMany lakes and parts of the ocean freeze over in the winter. Ice acts as a buffer between the atmosphere and the water, impacting the flux of heat and momentum, the exchange of gases, and the penetration of light. This has a profound impact on the movement of water and nutrients and its biological evolution. Reliable predictions of ice cover can ensure a more accurate simulation of the hydrodynamics, along with the nutrients, chemistry, and biology during the ice-covered months.\n\nHowever, ice cover is often partial, i.e., a fraction of the water body is covered with ice and the remainder is open water. Partial ice cover can exist for days, weeks, or even months, with the ice cover expanding and contracting as the weather changes. During such periods, some parts of the water body are exposed to the atmosphere while other parts of the water body have this ice cover buffer. This has both local effects on the water circulation and more general effects if partial ice cover exists for many days.\n\nCurrently, a hydrodynamic model is used to simulate the movement of water. To accurately simulate hydrodynamics during extended periods of partial ice cover, the hydrodynamics model needs knowledge of the spatial extent of ice cover.",
  "cpc": [
    "G01W 1/10",
    "B25J 9/1664",
    "G06N 20/00",
    "G06N 20/20",
    "G06N 5/01",
    "G16C 10/00",
    "G16C 20/10",
    "G16C 20/70"
  ],
  "ipc": [
    "B25J 9/16",
    "G01W 1/10",
    "G06N 20/00",
    "G06N 5/01"
  ],
  "assignees": [
    "International Business Machines Corp"
  ],
  "inventors": [
    "Campbell D. Watson",
    "Guillaume A. R. Auger",
    "Eli Michael Dow"
  ],
  "filing_date": "2020-12-01",
  "publication_date": "2025-09-30",
  "grant_date": "2025-09-30",
  "priority_date": "2020-12-01",
  "application_number": "US-202017108986-A",
  "family_id": "81751278",
  "cited_by_count": 0,
  "citations": [
    "US20110291862A1",
    "WO2013009245A1",
    "US20130099960A1",
    "US10168448B2",
    "US20140324352A1",
    "US10386541B2",
    "US20160167755A1",
    "CN104198052A",
    "KR101645468B1",
    "KR101603718B1",
    "US20200372359A1",
    "US10762422B2",
    "CN109196527A",
    "US20190228362A1",
    "EP3330747B1",
    "US20180252842A1",
    "JP2019087244A",
    "CN108375808A",
    "CN109117883A",
    "US20200208979A1",
    "CN109829569A",
    "US20200309993A1",
    "US20200348448A1",
    "US20200356839A1",
    "US20200372349A1",
    "CN111539842A",
    "CN114582440A",
    "JP2022087837A"
  ]
}