Search

EP-4435734-B1 - METHOD FOR ASSESSING CARBON CAPTURE OF AN AREA OF INTEREST

EP4435734B1EP 4435734 B1EP4435734 B1EP 4435734B1EP-4435734-B1

Inventors

  • WALLACE, ANDREW

Dates

Publication Date
20260513
Application Date
20230321

Claims (14)

  1. A computer-implemented method (20) for assessing carbon capture in an area of interest, the method comprising: - implementing (22) at least one image analysis algorithm on at least one top-down image of at least part of the area of interest, to determine environmental data representative of at least one primary producer and/or at least one biotope of the area of interest, said at least one image analysis algorithm comprising extracting, based on each top-down image, colour features comprising a normalized difference vegetation index and a normalized difference water index, and performing at least one of: ∘ image recognition to associate predetermined shapes and/or predetermined combinations of shapes and the extracted colour features in the retrieved top-down image with at least one primary producer and/or with at least one biotope; and ∘ applying an artificial intelligence model to analyse the top-down image, said artificial intelligence model having been previously trained based on said colour features in known areas where at least one primary producer has been clearly identified, to learn which colour features values are associated with said primary producer; - based on the determined environmental data, computing (24) a carbon capture indicator representative of an estimated carbon capture potential of the area of interest.
  2. The method (20) according to claim 1, wherein the computed carbon capture indicator is a net primary productivity of the area of interest.
  3. The method (20) according to claim 2, further comprising calculating the carbon capture potential of the area of interest based on the net primary productivity and a size of the area of interest.
  4. The method (20) according to any one of claims 1 to 3, wherein the determined environmental data include at least one primary producer of the area of interest, and wherein computing the carbon capture indicator includes associating each primary producer to a corresponding expected net primary productivity.
  5. The method (20) according to any one of claims 1 to 4, wherein the carbon capture indicator is further computed based on habitat data associated with the area of interest and representative of climatic features, pedological features and/or sediment characteristics, geological features, hydrographic features and/or topographic features of the area of interest.
  6. The method (20) according to any one of claims 1 to 5, wherein the environmental data is further determined based on a series of top-down images of the area of interest acquired at different acquisition dates.
  7. The method (20) according to any one of claims 1 to 6, further including retrieving on-site measurement data representative of at least one physical and/or chemical property of the area of interest, the carbon capture indicator being further computed based on the retrieved on-site measurement data.
  8. The method (20) according to any one of claims 1 to 7, wherein the step (22) of implementing at least one image analysis algorithm includes detecting the presence of at least one predetermined human-made structure in the area of interest, the carbon capture indicator being further computed based on each detected human-made structure.
  9. The method (20) according to any one of claims 1 to 8, further including predicting (26) an evolution of the carbon capture indicator over time based on the determined environmental data.
  10. The method (20) according to claim 9 when depending on claim 8, wherein the evolution of the carbon capture indicator over time is further predicted based on each detected human-made structure.
  11. The method (20) according to any one of claims 1 to 10, wherein at least one of the determined environmental data and the computed carbon capture indicator is associated with a date of acquisition of each corresponding top-down image, the method further including monitoring (28) an evolution of the environmental data and/or the carbon capture indicator over time, and outputting an alert signal if a corresponding variation over time is outside a predetermined range.
  12. The method (20) according to any one of claims 1 to 11, further comprising: - implementing at least one image analysis algorithm on at least one top-down image of at least one neighbouring area adjacent to the area of interest, to determine environmental data representative of at least one primary producer and/or at least one biotope of each neighbouring area; - for each neighbouring area, computing (30) a carbon capture indicator representative of an estimated carbon capture potential of said neighbouring area, based on the determined environmental data of said neighbouring area and on the computed carbon capture indicator of the area of interest.
  13. A computer program comprising instructions, which when executed by a computer, cause the computer to carry out the steps of the method of any one of claims 1 to 12.
  14. A remote monitoring system (2) for assessing carbon capture in an area of interest, the remote monitoring system (2) including a processing unit (6) configured to: - implement at least one image analysis algorithm on at least one top-down image of at least part of the area of interest, to determine environmental data representative of at least one primary producer and/or at least one biotope of the area of interest, said at least one image analysis algorithm comprising extracting, based on each top-down image, colour features comprising a normalized difference vegetation index and a normalized difference water index, and performing at least one of: ∘ image recognition to associate predetermined shapes and/or predetermined combinations of shapes and the extracted colour features in the retrieved top-down image with at least one primary producer and/or with at least one biotope; and ∘ applying an artificial intelligence model to analyse the top-down image, said artificial intelligence model having been previously trained based on said colour features in known areas where at least one primary producer has been clearly identified, to learn which colour features values are associated with said primary producer; - based on the determined environmental data, compute a carbon capture indicator representative of an estimated carbon capture potential of the area of interest.

Description

Field of the invention The present invention relates to a method for assessing carbon capture in an area of interest. The invention further relates to a computer program, and to a device configured to perform said method. The invention applies to the field of remote monitoring, and more specifically to the remote assessment of carbon capture of an area of interest. Background According to the 2021 Intergovernmental Panel on Climate Change (IPCC) report, it is required to limit global warming to 1.5°C above pre-industrial levels in order to prevent the most dangerous and irreversible aspects of climate change. In this context, atmospheric carbon dioxide, which is known to be a major contributor to greenhouse effect, has seen its levels reach 421 ppm (parts per million) in 2022, an increase, according to the National Oceanic and Atmospheric Administration (NOAA), of 50% over pre-industrial levels. Oceans cover 71% of the world's surface area, and research shows that marine primary producers have enormous potential as a source of carbon capture. Today, to measure the carbon capture potential of a given area of coastal or marine environment, it is required to perform on-site observations and measurements to obtain relevant data, and to determine the carbon capture of said area based on the obtained data. However, this method is not entirely satisfactory. Indeed, the obtained data are limited to the areas that can actually be reached by a human expert. Furthermore, such method is heavily based on human expertise, so that it lacks reliability, consistency and repeatability. A purpose of the present invention is to overcome at least one of these drawbacks. Another purpose of the invention is to provide a reliable, consistent and repeatable way to remotely measure the carbon capture potential of an area, especially an area of coastal or marine environment. Another purpose of the invention is to provide a method that does not rely as much on human expertise to perform such measurement. Cory C Cleveland et al: "A comparison of plot-based satellite and Earth system model estimates of tropical forest net primary production", Global Biochemical Cycles, American Geophysical Union, vol. 29, issue 5, pages 626-644, describes radiation-based net primary production estimates. Forrest G Hall et al: "Multi-angle narrow-band remote sensing of gross primary production", 2016 IEEE International Geoscience and Remote Sensing Symposium, IEEE, pages 1717-1718, describes multi-angle narrow-band remote sensing of gross primary production. US 2005/273358 A1 describes generating standardized environmental benefit credits. Summary of the invention To this end, the present invention is a method of the aforementioned type, comprising: implementing at least one image analysis algorithm on at least one top-down image of at least part of the area of interest, to determine environmental data representative of at least one primary producer and/or at least one biotope of the area of interest;based on the determined environmental data, computing a carbon capture indicator representative of an estimated carbon capture potential of the area of interest. Indeed, by determining the carbon capture potential of the area of interest based on top-down images of said area of interest, remote monitoring is achieved. In other words, the need to perform on-site measurement, especially in hard-to-reach and/or remote areas such as areas of coastal or marine environment, is overcome. Moreover, the outputs provided by the image analysis algorithm are repeatable and quantifiable. Consequently, the invention provides a reliable, consistent and repeatable way to remotely measure the carbon capture potential of an area of interest. According to other advantageous aspects of the invention, the method includes one or several of the following features, taken alone or in any technically possible combination: the computed carbon capture indicator is a net primary productivity of the area of interest;the method further comprises calculating the carbon capture potential of the area of interest based on the net primary productivity and a size of the area of interest;the determined environmental data include at least one primary producer of the area of interest, and wherein computing the carbon capture indicator includes associating each primary producer to a corresponding expected net primary productivity;the carbon capture indicator is further computed based on habitat data associated with the area of interest and representative of climatic features, pedological features and/or sediment characteristics, geological features, hydrographic features and/or topographic features of the area of interest;the step of implementing at least one image analysis algorithm includes computing, based on each top-down image, a normalized difference vegetation index and a normalized difference water index, the environmental data being determined based on the computed normalized difference ve