Assessing carbon sequestration contributions of nature-based assets located in aquatic environments

Abstract

A method, computer program product, and computer system for assessing carbon sequestration contribution of a nature-based asset located in a water environment are provided. The method includes obtaining a meta descriptor of the nature-based asset and a meta descriptor of a region of the water environment in which the asset is located; mapping a size of the nature-based asset; accessing data of a monitored physical property of the asset in the water environment based on the meta descriptor of the asset; accessing data of a monitored carbon concentration in the region of the water environment based on the meta descriptor of the region; accessing data of monitored physical, chemical, and biological properties of the water environment based on the meta descriptor of the region; and applying a model to the accessed data to assign a carbon sequestration contribution per unit of measurement of the nature-based asset.

Description

Background Technology

[0001] This invention relates to carbon sequestration, and more specifically, to assessing the carbon sequestration contribution of nature-based assets located in aquatic environments.

[0002] The world's oceans are the largest continuous carbon sink for releasing carbon dioxide from the atmosphere. Current estimates put approximately 1.8 Pg C into the ocean each year, but the exact figure remains highly uncertain (approximately 0.7 Pg C per year).

[0003] Different regions of the ocean are not equally effective at absorbing CO2. Some regions (such as the equatorial Pacific) release CO2 into the atmosphere, while others (such as the North Atlantic) absorb it. In addition to this regional difference, there can also be significant temporal differences.

[0004] Estimates suggest that nature-based solutions could provide 37% of the emissions reductions needed to achieve the Paris Agreement targets by 2030. However, current solutions are uncertain, lack transparency, and are vulnerable to malicious actors.

[0005] The carbon market is a trading system for buying and selling carbon credits to offset greenhouse gas emissions. In the near to medium term, the carbon market is expected to be subject to stricter regulation to introduce greater consistency, strengthen the integrity of sustainability disclosures, and respond to stakeholder expectations that sustainability information should be transparent and comparable. Examples of these regulatory processes include the outputs of the Working Group on Disclosure of Nature-Related Financial Information and the Working Group on Disclosure of Climate-Related Financial Information.

[0006] Currently, owners or stakeholders of nature-based assets must register with a verification agency and create or register new projects describing carbon credits. Project documents are uploaded, describing details such as project name, project size, location, and the average annual amount of voluntary carbon units (1 unit = 1 tonne of CO2). A comprehensive review of the project documents is conducted, typically involving two to three rounds of comments and responses. Once approved by the registration agency, the owner or stakeholder can sell voluntary carbon units. Summary of the Invention

[0007] According to one aspect of the invention, a computer-implemented method is provided for assessing the carbon sequestration contribution of a nature-based asset located in an aquatic environment, the method comprising: obtaining a meta-descriptor of the nature-based asset and a meta-descriptor of a region of the aquatic environment in which the asset is located; mapping the size of the nature-based asset; accessing data on monitored physical properties of the asset in the aquatic environment based on the asset's meta-descriptor; accessing data on monitored carbon concentrations in the region of the aquatic environment based on the region's meta-descriptor; accessing data on monitored physical, chemical, and biological properties of the aquatic environment based on the region's meta-descriptor; and applying a model to the accessed data to assign a per-unit measured carbon sequestration contribution of the nature-based asset.

[0008] The advantage of this approach lies in its ability to validate the carbon sequestration contribution of nature-based assets by modeling data collected from the assets and their aquatic environment. This provides a precise metric for carbon absorption and storage, crucial for achieving the net-zero target. The approach uses direct measurements rather than surrogate measurements to provide robust and defensible quantitative indicators.

[0009] According to another aspect of the invention, a system for assessing the carbon sequestration contribution of a nature-based asset located in an aquatic environment is provided, comprising a processor and a memory configured to provide computer program instructions to the processor to perform the functions of the following components: a meta-descriptor component for obtaining meta-descriptors of the nature-based asset and meta-descriptors of a region of the aquatic environment in which the asset is located; a mapping component for mapping the size of the nature-based asset; an asset data monitoring component for accessing data on monitored physical properties of the nature-based asset in the aquatic environment based on the meta-descriptors; a carbon concentration data component for accessing data on monitored carbon concentrations in a region of the aquatic environment based on the meta-descriptors; an aquatic environment data component for accessing data on monitored physical, chemical, and biological properties of the aquatic environment based on the meta-descriptors of the region; and a model application component for applying a model to assign a per-unit measured carbon sequestration contribution of the nature-based asset.

[0010] According to another aspect of the invention, a computer program product is provided for assessing the carbon sequestration contribution of a nature-based asset in an aquatic environment. The computer program product includes a computer-readable storage medium having program instructions executable by a processor to cause the processor to: obtain a meta-descriptor of the nature-based asset and a meta-descriptor of a region in the aquatic environment in which the asset is located; map the size of the nature-based asset; access data on monitored physical properties of the nature-based asset in the aquatic environment based on the asset's meta-descriptor; access data on monitored carbon concentrations in the region of the aquatic environment based on the region's meta-descriptor; access data on monitored physical, chemical, and biological properties of the aquatic environment based on the region's meta-descriptor; and apply a model to the accessed data to assign a per-unit measured carbon sequestration contribution of the nature-based asset. The computer-readable storage medium may be a non-transient computer-readable storage medium, and the computer-readable program code may be executable by processing circuitry. Attached Figure Description

[0011] Embodiments of the invention will now be described by way of example only, and with reference to the accompanying drawings: Figure 1 This is a flowchart of an example embodiment of a method according to an embodiment of the present invention; Figure 2 This is a block diagram of an example embodiment of a system according to an embodiment of the present invention; Figure 3 This is a flowchart of an example embodiment of the training method according to an embodiment of the present invention; and Figure 4 This is a block diagram of an example embodiment of a computing environment for executing at least some of the computer code involved in the present invention.

[0012] It should be understood that, for simplicity and clarity, the elements shown in the accompanying drawings are not necessarily drawn to scale. For example, some elements may be enlarged relative to others for clarity. Furthermore, reference numerals may be repeated in the accompanying drawings where appropriate to indicate corresponding or similar features. Detailed Implementation

[0013] Examples of methods, systems, and computer program products for assessing the carbon sequestration and other contributions of aquatic nature-based assets are provided. Aquatic nature-based assets are natural or man-made resources that generate positive environmental value. Examples include natural or man-made coral reefs, mangroves, seagrass farms, oyster habitats, etc. Nature-based assets can also be referred to as nature-based solutions because they can be developed to address carbon emissions issues.

[0014] The described methods and systems use models, based on model inputs, to assign a per-unit measurement (e.g., per square unit area or per cubic unit volume) carbon sequestration contribution to natural-based assets located in water. The model can be a trained artificial intelligence modeling system (e.g., a transformer model network, graph neural network, seq2seq network, etc.). In other embodiments, the modeling system can be a rule-based model.

[0015] The described methods and systems involve accessing data on monitored carbon concentrations in a region of an aquatic environment (e.g., ocean or other water bodies) using a sensing platform that may include artificial intelligence sensing devices; accessing data on monitored physical properties of natural-based assets in the aquatic environment; accessing data on monitored physical, chemical, and biological properties of the aquatic environment; and mapping the size of the natural-based assets. This data is fed into a modeling system to assign a per-unit measured carbon sequestration contribution to the natural-based assets.

[0016] The output is an assessment of nature-based assets. This can include metrics such as carbon sequestration, biodiversity, the asset's natural contribution, the physical extent and characteristics of the nature-based asset, and the health or state index of the nature-based asset.

[0017] This method and system can use an immutable data repository (e.g., a blockchain) to store nature-based assets, along with associated time-varying carbon sequestration and / or other nature-based contributions, in the form of tokens. Such storage can inject trust and transparency into the system.

[0018] The described methods and systems can be applied to combine observation, machine learning, and immutable data repositories to validate and verify nature-based solutions located in water (i.e., ocean or freshwater) and tokenize nature-based contributions in voluntary or compliant markets. This can provide a framework to distinguish baseline carbon sequestration from contributions from specific nature-based assets located in water, such as coral reefs (natural or man-made) or mangroves. The methods and systems can simultaneously monitor the state or health of nature-based assets.

[0019] Assessing the carbon sequestration contribution of nature-based assets located in aquatic environments is an improvement in the field of environmental monitoring technology, particularly in the field of environmental monitoring and assessment for carbon sequestration purposes.

[0020] The following terms are used in this instruction manual.

[0021] Carbon sequestration is the process of capturing, fixing, and storing carbon dioxide from the atmosphere.

[0022] A “voluntary carbon market” allows carbon emitters to offset their emissions by purchasing carbon credits issued by projects designed to remove greenhouse gases from the atmosphere or reduce their emissions.

[0023] A “compliant carbon market” is the result of any national, regional, and / or international policy or regulatory requirements.

[0024] The European Commission defines “nature-based solutions” as “nature-inspired and supported solutions that are cost-effective while providing environmental, social and economic benefits and contributing to resilience. Such solutions bring more and more diverse nature and natural features and processes into urban, terrestrial and marine landscapes through site-specific, resource-efficient and systematic interventions.”

[0025] A carbon market is a system where countries, companies, or individuals can buy and sell carbon emission allowances or credits to meet their emissions reduction targets. Key stakeholders include governments that set emissions regulations and caps; companies that emit greenhouse gases (and therefore need to buy credits) or reduce emissions (and can sell credits); and intermediaries or brokers who facilitate the trading process. The market operates on the principle of supply and demand: entities that need to offset emissions buy credits from entities with excess emissions reductions, generating revenue for sellers who can provide verifiable emissions reductions at below-market prices.

[0026] "Validation" and "verification" are crucial steps in a nature-based project. During "validation," the accreditation body determines whether the project meets all the rules and requirements of the program. Upon successful validation, the project supporter can submit the project to the relevant program for registration. During "verification," the accreditation body confirms that the deliverables specified in the project documentation have been achieved and quantified in accordance with the requirements of the relevant standards.

[0027] refer to Figure 1 Flowchart 100 illustrates an example embodiment of the described method for assessing the carbon sequestration contribution of nature-based assets.

[0028] This method may include training 101 artificial intelligence or building a rule-based model to assign a per-unit measured carbon sequestration contribution to nature-based assets in a regional aquatic environment. Training may include feature extraction, which may combine: learning relevant environmental features that influence carbon sequestration potential; estimating carbon concentration values ​​from hydrological and atmospheric environmental variables; estimating the contributions of different features to model predictions and their spatial and temporal characteristics; and extracting information about the physical and material properties of the assets.

[0029] This method can obtain 102 meta-descriptors for nature-based assets. This can include detailed information about the asset, the area to be assessed, the specific attributes of the introduced asset (e.g., geopolymers that absorb carbon), and information on the regulatory framework or standards used to report carbon sequestration contributions.

[0030] The method may include mapping the size of a nature-based asset 103. This may include the spatial size of the asset, specific attributes of the region (e.g., substrate conditions), and additional natural or anthropogenic characteristics (e.g., the presence or introduction of seagrass farms).

[0031] This method may include accessing data on the monitored physical properties of nature-based assets in the 104 aquatic environment. Monitoring the physical properties of nature-based assets in the area may include using acoustic and / or visual devices.

[0032] This method may include accessing data on monitored carbon concentrations in a 105-area water environment region. This can utilize sensing platforms and AI-enabled sensing devices. The method may also include accessing additional environmental information datasets of the 106-area water environment, including the monitored physical, chemical, and biological properties. Data on carbon concentrations in the monitored water environment region may include in-situ data and exogenous environmental data. Monitoring carbon concentrations in the water environment region can utilize in-situ data and remote sensing for rapid, autonomous monitoring of water variables.

[0033] Data collection strategies can be updated based on model output to improve the system’s ability to collect the most relevant data for model performance and carbon estimation.

[0034] This method employs 107 trained artificial intelligence (AI) or rule-based models to assign a per-unit measured carbon sequestration contribution to nature-based assets. The applied model can include meta-descriptors for the input nature-based assets and regions. Additional attributes can also be specified, such as the category(s) of the claimed nature-based solution (e.g., mangroves, seagrass farms, coral reefs, oyster habitats, etc.), attributes of the intervention (such as material composition), and applicable regulatory frameworks and standards. The model can also ingest data from IoT sensors, water quality sampling, numerical model analysis, remote sensing estimation, drone imagery, satellite imagery, and acoustic data.

[0035] This method can output 108 indicators of the contribution and state of nature-based assets. It can include applying models to process and classify health indices of nature-based assets. It can also include applying trained artificial intelligence models to predict the carbon sequestration of assets over time.

[0036] This method may include updating the 109 data collection strategy based on modeling location sensing needs to model the locations where sensor devices are deployed in a region. This includes increasing sampling in areas with high data variance based on statistical indicators of the monitored data. This can leverage AI components to dynamically optimize sensor deployment based on statistical indicators of the data or model training. Location modeling can be applied to autonomous surface or underwater vehicles or devices to deploy sensor devices in different locations. This may include an AI-guided sampling module that guides the vehicle's trajectory to optimize the sampling strategy (e.g., more intensive sampling in areas with high variance or areas that contribute significantly to model predictions).

[0037] The updated data collection strategy 109 can monitor the contribution of different features to the prediction and prioritize the collection of certain features based on feature importance or contribution to the prediction.

[0038] This method can store the tokenized contributions of nature-based assets in an immutable data repository. The token can define the asset and its associated carbon sequestration over time. Additional information stored in the immutable data repository can include data used to calculate the carbon sequestration contribution, as well as a framework for estimation.

[0039] refer to Figure 2 The block diagram illustrates system 200 as an example embodiment, including the described assessment system 210 for evaluating the carbon sequestration contribution of natural-based assets in water. In this example embodiment, a trained AI modeling system 232 is trained and applied. In other embodiments, a rule-based modeling system can be built and used.

[0040] Evaluation system 210 can be implemented on computing system 201, which may include at least one processor 202, hardware modules, or circuitry for performing the described component functions, which may be software units executing on at least one processor. Multiple processors running parallel processing threads may be provided to achieve parallel processing of some or all of the component functions. Memory 203 may be configured to provide computer instructions 204 to at least one processor 202 to perform the component functions. The illustrated system and components may be provided on multiple computing systems.

[0041] The evaluation system 210 includes an input component 220 for accessing monitored data, an AI model application component 230 using a trained modeling system 232, and an output component 250 for outputting modeling results.

[0042] The evaluation system 210 may have an associated model training system 290 for training a trained modeling system 232 associated with the evaluation system 210. The model training system 290 may include training an artificial intelligence model to assign per-unit measured carbon sequestration contribution of a natural asset by using a combination of feature extractions from: carbon concentration estimates and driving factors from water and / or atmospheric environmental data; and fundamental asset variables from asset monitoring data. Training may also include asset characteristics provided as input, as well as regulatory frameworks and standards.

[0043] The evaluation system 210 may include a meta-descriptor component 225 for providing parameters for assets and regions. The input component 220 may include a meta-descriptor input component 226 for inputting meta-descriptors of assets into a trained artificial intelligence model. The input component 220 may include a mapping component 224 for mapping natural-based asset dimensions.

[0044] Input component 220 may include asset data monitoring component 221 for accessing monitored physical properties of natural assets in the aquatic environment based on meta-descriptors. Input component 220 may include carbon concentration data component 222 for accessing monitored carbon concentration data in aquatic environment areas based on meta-descriptors. Input component 220 may include aquatic environment data component 223 for accessing monitored physical, chemical, and biological properties of the aquatic environment based on region meta-descriptors.

[0045] The evaluation system 210 may include a data collection update component 291 for updating the data collection strategy based on the performance of model training.

[0046] The evaluation system 210 may also include a tokenization system 240 for storing tokens related to modeling and evaluation in an immutable data storage repository 242. The immutable data repository 242 may store nature-based assets and their associated time-varying carbon sequestration or other nature-based contributions in token form.

[0047] The assessment system 210 can be used to assess nature-based assets 282 in an aquatic environment area 280 (such as a marine or freshwater area). A trained modeling system 232 can be trained and applied to the input to assign a per-unit (e.g., per square meter) carbon sequestration contribution or other nature-based contributions to the nature-based assets. The AI ​​model application component 230 may include a prediction component 234 for applying the trained modeling system 232 to predict the carbon sequestration of the assets over time.

[0048] The autonomous vehicle 270 can be used to acquire data related to the nature-based asset 282 and the aquatic environment area 280, and it can communicate with the assessment system 210. A carbon concentration monitoring system 273 and a nature-based asset monitoring system 272 can be provided for data collection in the area. The autonomous vehicle 270 can be an autonomous surface or underwater vehicle or device that can deploy sensor equipment from different locations. The autonomous vehicle 270 may include edge computing components for deploying sensors and collecting and transmitting data.

[0049] The autonomous vehicle 270 may also include a sensing position system 271. The sensing position system 271 may include an AI-guided sampling module that guides the vehicle's trajectory to optimize sampling strategies (e.g., more intensive sampling in areas with high variance). The sensing position system 271 may include a sensing position optimization system for quantifying the contribution of different features or subsets of features to model predictions and using this information to guide sampling location and frequency. This may include increasing sampling capacity in areas with high data variance or in areas that contribute more significantly to AI model training based on statistical indicators of the monitored data.

[0050] Carbon concentration monitoring system 273 can be used to monitor carbon concentration in aquatic environments by rapidly and autonomously monitoring water variables using in-situ data and remote sensing. Carbon concentration monitoring system 273 may include a sensing platform and / or AI-enabled sensing devices capable of rapidly and autonomously measuring carbon concentration in water, as well as optional other parameters. Carbon concentration monitoring system 273 can be deployed on fixed equipment (e.g., a buoy equipped with a vertical sampler) that can measure chemical and physical properties over time. The carbon concentration monitoring system can sample other water and atmospheric variables and use AI or rule-based models to estimate carbon concentration values ​​based on these variables.

[0051] The nature-based asset monitoring system 272 may include acoustic and / or computer vision devices that can be used to monitor the physical characteristics of nature-based assets, such as coral reefs, mangrove reefs, and seagrass forests.

[0052] The system may also include a remote sensing system 260 module for remotely sensing relevant ocean color variables (e.g., temperature, salinity, chlorophyll a). The remote sensing system 260 may be a satellite sensing system for geospatial measurements of ocean or other water color variables to provide spatial estimates of surface measurements at daily, weekly, or monthly frequencies.

[0053] The trained modeling system 232 provides the carbon sequestration contribution per square meter of nature-based assets or other nature-based contributions based on measurements sampled by the described system. The trained modeling system 232 provides AI-based processing of nature-based health indices, carbon sequestration contributions, and the trajectory or evolution of nature-based assets.

[0054] Modeling System

[0055] The input to an applied modeling system may include at least some of the following data: Parameters of the area to be monitored; Meta-descriptors (type, age, size, status) of natural assets; Optional additional features (ocean model data, weather data); and Detailed information on the regulatory framework or standards related to the carbon market or trading system, which will provide contributions based on nature.

[0056] Meta-descriptors for nature-based assets can include location, planned area, asset type (e.g., coral reef, mangrove reef, seagrass, etc.), age or maturity level, state, or any other descriptor. These can be used as static attributes for training artificial intelligence models.

[0057] The modeling system can access monitored data within a modeled area related to carbon concentration, which could be point measurements of carbon dioxide saturation (pCO2) at different depths in the ocean using chemical sensors. Chemical sensors can include AI-assisted chemical sensing and software-defined sensors, such as HyperTaste (a trademark of IBM).

[0058] The modeling system can access monitored data in the modeled area associated with the nature-based asset, such as measurements of the asset's primary productivity using chlorophyll a sensors or "AI microscopes" (small autonomous microscopes that can be placed in water to monitor plankton in situ), and can identify different species and track their movement in three-dimensional space.

[0059] These monitored datasets can be extended using other commonly used ocean or hydrological datasets, such as conductivity, temperature, dissolved oxygen, pH, and turbidity. Additional geospatial datasets, such as remote sensing data, weather information, and public ocean model data, can also be used.

[0060] Using sonar or equivalent underwater acoustic detection capabilities, precise asset dimensions can be mapped through wavelength conversion, enabling continuous quantification of reef maturity measurements, as well as the status of vegetation, fauna, and other features. Alternatively, asset size and characteristics can also be obtained through manual sampling or measurement.

[0061] A trained machine learning modeling system can be provided to assign a nature-based contribution per unit of measurement. This could include carbon sequestration or other environmental or ecosystem contributions. The system can leverage pre-trained base models employing techniques such as transformer architectures and pre-trained across diverse datasets.

[0062] Pre-trained AI models process environmental indicator data from sensors, reef physical indicator data, and any other datasets to generate indicators such as: carbon sequestration, biodiversity, nature-based contribution of assets; physical extent and characteristics of reefs; and / or reef health or status indices. This can include the ability to update the model using online learning capabilities. It can include the ability to provide textual prompts to the model as input. It can include multimodal capabilities, allowing the model to ingest data from diverse data types, such as time series, text, images, graphs, and point clouds.

[0063] The output can be tokenized in an immutable data repository (e.g., a blockchain) as a verified asset that can be sold on a voluntary carbon market.

[0064] The modeling system can provide at least some of the following outputs: Carbon sequestration per unit area per unit time (or other nature-based contributions, such as biodiversity). Background or baseline levels of carbon sequestration, water quality, or ecosystem services; Changes or time-series evolution of carbon sequestration, water quality, or ecosystem services; The physical scope and characteristics of the assets; Based on natural asset health indices or status; and Changes or modifications to asset characteristics and performance.

[0065] Outputs can include quantifiable metrics of nature-based contributions per unit area or per unit time (e.g., per square meter per day). Quantifiable metrics can include: the amount of carbon absorbed, contributions to improving biodiversity or ecosystem health, or contributions to improving coastal resilience or other infrastructure.

[0066] Outputs may include indicators that quantify the health or status of a nature-based asset, which could include a trajectory or evolution of asset performance, or an early indicator of potential health deterioration (and the subsequent potential reduction in nature-based contributions). This could be achieved using specific water quality indices, such as the Coastal Eutrophication Index (ICEP) or the National Sanitation Foundation Water Quality Index (NSF-WQI). Alternatively, appropriate alternative indicators may be used.

[0067] The output can be tokenized in an immutable data repository (e.g., a blockchain) as a verified asset that can be sold on a voluntary carbon market.

[0068] This can be used to tokenize nature-based contributions within a blockchain network. Tokenized information may include: details of the nature-based asset and its meta-descriptor; details of the baseline level of carbon sequestration or other nature-based contributions; details of changes in the level of carbon sequestration or other nature-based contributions; data used to quantify these contributions; and data on the standards or indices used.

[0069] Environmental monitoring and sensing

[0070] Monitoring data related to carbon concentration can be obtained by measuring carbon dioxide saturation (pCO2) points at different depths in the ocean using chemical sensors. Chemical sensors can include AI-assisted chemical sensing utilizing software-defined sensors, such as HyperTaste (a trademark of IBM). Potentiometric electronic tongue technology combines miniaturization, edge computing, and AI for the chemical analysis of liquids. A micro-controlled data acquisition program is combined with a miniaturized array of multiple electrodeposited conductive polymers. The data is fed into an AI model to identify chemical components.

[0071] The modeling system can access relevant monitoring data on nature-based assets in the modeled area by measuring the primary productivity of the assets, such as using chlorophyll-a sensors or “AI microscopes” (small autonomous microscopes that can be placed in the water) for in-situ monitoring of plankton, identifying different species and tracking their three-dimensional movement. This makes it easy to track the ever-evolving biological properties of the ocean, especially changes in biodiversity.

[0072] Monitoring data related to the environmental condition and ecosystem services of assets can be collected by IoT or other sensing devices that measure different points in the environment. These devices can collect data such as conductivity, temperature, dissolved oxygen, turbidity, and pH.

[0073] Data on environmental conditions, carbon sequestration, and nature-based assets can be collected using autonomous or remotely operated surface or underwater vehicles or devices. The trajectories of these devices, along with the associated sampling locations and time periods, can be guided by AI models based on contributions to model training or statistical variance or descriptors of the data.

[0074] Physical asset sensing

[0075] This disclosure collects information about the characteristics, scope, health, and conditions of nature-based assets in order to assign nature-based contributions to specific assets. Example monitoring may include the following techniques.

[0076] Computer vision data can be used to collect specific data on the health of nature-based assets and can be combined with AI microscopy to provide highly detailed measurements in specific situations. This can leverage hyperspectral vision. Echo sounder technology can be used to generate high-resolution 3D topographic maps of aquatic areas, such as the seabed. These technologies can be used to continuously and autonomously map the extent and growth of assets, such as coral reefs, for example, on an underwater glider.

[0077] This data can also be used to extract indicators of asset health. For example, for assets in the form of coral reefs, this can be based on the proven predictive power of the principle that "a noisy reef is a healthy reef." This can extend research from shipboard surveys to monitoring underwater acoustics.

[0078] Asset sensing systems can generate autonomous, continuous data that characterizes the spatial extent, growth, health, and biological processes of assets.

[0079] Monitoring can utilize autonomous vehicles and gliders to extend spatial coverage to nature-based assets of any size.

[0080] Remote sensing technologies, including satellite or airborne measurements, can be used to measure surface variables (temperature, salinity, chlorophyll-a) on a large scale. Quantification of ocean carbon dioxide sequestration can be used to improve quantification methods.

[0081] Data from other sources, such as weather data and ocean model data, can also be used.

[0082] AI Model Training and Feature Extraction

[0083] Data from environmental and asset sensors are used to train an AI model to classify the nature-based contributions of assets (e.g., carbon sequestration, biodiversity). The AI ​​model processes point measurements of environmental indicators (CO2, dissolved oxygen, temperature, chlorophyll-a, pH, etc.) and geospatial (location-specific) indicators (sea surface temperature, salinity, chlorophyll-a, etc.) and creates a quantification of the nature-based contribution per square meter of seabed.

[0084] The model's input can include multimodal datasets, such as time series, geospatial or grid data, images, text, or point cloud data. This data can be ingested using a pre-trained base model, encoder-decoder architecture, autoencoder, graph neural network, or transformer-based architecture to adapt to the dataset's complexity and train the model with an appropriate feature set. Automatic AI (Auto-AI) methods can also be used for feature extraction, which employ grid search to process the data, extracting the most relevant features and identifying possible transformations of these features to improve the model's predictive capabilities.

[0085] Data from environmental and reef sensors are also used to train AI models to classify asset health status. Model features can include visual, acoustic, and textual data, as well as other physical descriptors of the assets based on their natural characteristics and corresponding health indices.

[0086] AI models can be trained in a supervised manner with appropriately labeled data, or in an unsupervised manner, such as using masked self-attention.

[0087] Figure 3 Schematic diagram 300 is shown, illustrating an example embodiment of feature extraction in an AI modeling system.

[0088] Access in-situ sensor measurements 301, which may include: temperature, pH, DO, salinity, chlorophyll-a, and pCO2.

[0089] Accessing external environmental measurements 302 may include: 1. Remote sensing of SST, salinity, chlorophyll-a; and 2. Numerical oceanographic or atmospheric product data; and 3. Unstructured scientific articles or reports about the state of the environment.

[0090] Accessing physical asset monitoring metrics 303 may include: 1. Visual or camera data; 2. Underwater acoustic sampling, such as echo sounders; 3. High-resolution images, such as hyperspectral images; and 4. Natural language descriptors for assets.

[0091] The in-situ sensor measurements 301 and the external environmental measurements 302 can be combined by an AI model and converted 310 into carbon estimates to extract carbon concentration estimates 311 and their related driving factors.

[0092] For example, carbon concentration estimation 311 and its associated driving factors may include: pCO2 estimation and driving factors such as temperature, dissolved oxygen, and pH.

[0093] Physical asset monitoring indicators 303 can be used to extract basic asset variables 312. These may include spatial extent and size, specific natural restoration assets (such as seagrass farms, seaweed or coral reefs), and specific carbon sequestration geopolymers deployed as part of the asset.

[0094] For example, if the physical asset is a coral reef, asset basic variable 312 could include: 1. Coverage and composition of reef-building corals; 2. The coverage rate of large algal canopies; and 3. Fish diversity and abundance.

[0095] Potentially relevant features 320 can be extracted from carbon concentration estimates 311 and their drivers, as well as fundamental asset variables 312. This can include manual feature extraction steps to extract relevant features, or feeding all the data into the model and using methods such as self-attention to learn the relevant features.

[0096] Potential relevant features 320 may include: • Time series data: CO2 concentration, environmental variables, asset health index. • Meta-information: Asset location, including latitude, longitude, and depth; time; asset extent; physical properties of the asset, such as vegetation cover; specific remediation activities introduced, such as algae growth or oyster beds; and any other nature-based contributions, such as carbon sequestration geopolymers or enhanced remediation efforts.

[0097] AI modeling system 330 estimates the nature-based contribution of assets. For example, for assets in the form of coral reefs, this could include carbon sequestration based on the reef's state. It could also include the biodiversity contribution of coral reefs as outlined in the Kunming-Montreal Global Biodiversity Framework.

[0098] Implementation examples may include regression model 331 for predicting carbon sequestration given asset status and associated environmental status.

[0099] The described methods and systems leverage AI and edge-based data collection to generate reliable estimates of nature-based contributions from diverse aquatic environmental assets. This utilizes AI to extract contributions attributable to specific assets rather than baseline environmental processes. The methods and systems are adaptable to different spatial and temporal scales. This provides a transparent and secure framework for the verification, tokenization, and sale of nature-based contributions, such as voluntary carbon units.

[0100] This method and system take into account the number of sensor measurements required to train a model over a vast ocean area. This includes an AI system that evaluates model estimates and updates the spatial and temporal frequencies of sensor measurements to optimize model performance.

[0101] The measurement, estimation, and analysis of ocean processes are highly sensitive to spatial and temporal variations. Insufficient spatial coverage limits model performance. Introducing AI-enabled systems to dynamically optimize sensor deployment can improve model performance. This can include monitoring the statistical properties of the collected data and prioritizing areas with high variance, or monitoring the training or fine-tuning of the AI ​​model and prioritizing areas that contribute more to model training.

[0102] Tokenizing metrics of asset or product characteristics or descriptors, as well as measurement and modeling data on carbon sequestration contributions, within an immutable data repository helps enhance the authenticity and credibility of carbon credit contributions.

[0103] By using direct measurement and AI (rather than alternative metrics), this disclosure provides a robust and defensible quantitative metric that meets current and future regulatory requirements.

[0104] Various aspects of this disclosure are described by narrative text, flowcharts, block diagrams of computer systems, and / or machine logic block diagrams included in embodiments of a computer program product (CPP). For any flowchart, depending on the art involved, operations may be performed in a different order than those shown in a given flowchart. For example, again according to the art involved, two operations shown in consecutive flowchart blocks may be performed in reverse order, as a single integrated step, concurrently, or in a manner that at least partially overlaps in time.

[0105] Computer program product embodiment (“CPP embodiment” or “CPP”) is a term used in this disclosure to describe a collection of one or more storage media (also referred to as “media”) included in a collection of one or more storage devices that collectively include machine-readable code corresponding to instructions and / or data for performing computer operations specified in a given CPP claim. A “storage device” is any tangible device capable of holding and storing instructions for use by a computer processor. Without limiting the foregoing, a computer-readable storage medium can be an electronic storage medium, a magnetic storage medium, an optical storage medium, an electromagnetic storage medium, a semiconductor storage medium, a mechanical storage medium, or any suitable combination of the foregoing. Some known types of storage devices that include these media include: floppy disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), optical disc read-only memory (CD-ROM), digital versatile optical disc (DVD), memory sticks, floppy disks, mechanical encoding devices (such as punched cards or pits / planes formed on the main surface of a disc), or any suitable combination of the foregoing. The term "computer-readable storage medium" as used in this disclosure should not be construed as storage itself being a transient signal, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides, optical pulses transmitted through fiber optic cables, electrical signals transmitted through wires, and / or other transmission media. As those skilled in the art will understand, during normal operation of a storage device, data is typically moved at certain points in time (such as during access, defragmentation, or garbage collection), but this does not make the storage device transient, because data is not transient when it is stored.

[0106] refer to Figure 4The computing environment 400 includes examples of environments for executing at least some of the computer code (such as evaluation system code 450) involved in implementing the methods of the present invention. In addition to box 450, the computing environment 400 also includes, for example, a computer 401, a wide area network (WAN) 402, an end-user equipment (EUD) 403, a remote server 404, a public cloud 405, and a private cloud 406. In this embodiment, the computer 401 includes a processor set 410 (including processing circuitry 420 and a cache 421), a communication structure 411, volatile memory 412, persistent storage 413 (including an operating system 422 and box 450 identified above), a peripheral device set 414 (including a user interface (UI) device set 423, a storage device 424, and an Internet of Things (IoT) sensor set 425), and a network module 415. The remote server 404 includes a remote database 430. Public cloud 405 includes gateway 440, cloud orchestration module 441, host physical machine set 442, virtual machine set 443, and container set 444.

[0107] Computer 401 can take the form of a desktop computer, laptop computer, tablet computer, smartphone, smartwatch or other wearable computer, mainframe computer, quantum computer, or any other computer or mobile device now known or developed in the future capable of running programs, accessing networks, or querying databases (such as remote database 430). As is well known in the field of computer technology, and depending on the specific technology, the execution of computer-implemented methods can be distributed across multiple computers and / or multiple locations. On the other hand, in this presentation of computing environment 400, for the sake of simplicity, the detailed discussion focuses on a single computer (specifically computer 401). Computer 401 can reside in the cloud, even... Figure 4 It is not shown in the cloud. On the other hand, unless there is any explicit instruction, computer 401 is not required to be in the cloud.

[0108] Processor assembly 410 includes one or more computer processors of any type, now known or to be developed in the future. Processing circuitry 420 may be distributed across multiple packages, for example, multiple cooperating integrated circuit chips. Processing circuitry 420 may implement multiple processor threads and / or multiple processor cores. Cache 421 is memory located within the processor chip package(s) and is typically used to store data or code that should be readily accessible by the threads or cores running on processor assembly 410. Cache memory is typically organized into multiple tiers based on its relative proximity to the processing circuitry. Alternatively, some or all of the caches in the processor assembly may be located “off-chip.” In some computing environments, processor assembly 410 may be designed to process qubits and perform quantum computing.

[0109] Computer-readable program instructions are typically loaded onto computer 401 to cause the processor set 410 of computer 401 to perform a series of operational steps to implement a computer-implemented method, such that the executed instructions instantiate the method specified in the flowcharts and / or narrative descriptions of the computer-implemented method included in this document (collectively, the “method of the invention”). These computer-readable program instructions are stored in various types of computer-readable storage media, such as cache 421 and other storage media discussed below. The processor set 410 accesses the program instructions and associated data to control and direct the execution of the method of the invention. In computing environment 400, at least some of the instructions for performing the method of the invention may be stored in block 450 of persistent storage device 413.

[0110] Communication structure 411 is a signal transmission path that allows the various components of computer 401 to communicate with each other. Typically, this structure consists of switches and conductive paths, such as switches and conductive paths that form buses, bridges, physical input / output ports, etc. Other types of signal communication paths, such as fiber optic communication paths and / or wireless communication paths, may also be used.

[0111] Volatile memory 412 is any type of volatile memory now known or developed in the future. Examples include dynamic random access memory (RAM) or static RAM. Typically, volatile memory 412 is characterized by random access, but is not required unless explicitly indicated. In computer 401, volatile memory 412 is located in a single package and inside computer 401, but alternatively or additionally, volatile memory may be distributed across multiple packages and / or located outside computer 401.

[0112] The persistent storage device 413 is any form of non-volatile storage device for a computer, now known or to be developed in the future. The non-volatility of this storage device means that the stored data is retained regardless of whether power is supplied to the computer 401 and / or directly to the persistent storage device 413. The persistent storage device 413 may be a read-only memory (ROM), but typically at least a portion of the persistent storage device allows data to be written, deleted, and rewritten. Some common forms of persistent storage devices include hard disks and solid-state storage devices. The operating system 422 can take many forms, such as various known proprietary operating systems or operating systems employing an open-source portable operating system interface type with a kernel. The code included in box 450 typically includes at least some of the computer code involved in performing the methods of the present invention.

[0113] Peripheral device set 414 includes a set of peripheral devices for computer 401. Data communication connections between peripheral devices and other components of computer 401 can be implemented in various ways, such as Bluetooth connections, near field communication (NFC) connections, connections via cables (such as Universal Serial Bus (USB) type cables), plug-in connections (e.g., secure digital (SD) cards), connections via local area network communication networks, and even connections via wide area networks (such as the Internet). In various embodiments, UI device set 423 may include components such as displays, speakers, microphones, wearable devices (such as goggles and smartwatches), keyboards, mice, printers, touchpads, game controllers, and haptic devices. Storage device 424 is an external storage device (such as an external hard drive) or a pluggable storage device (such as an SD card). Storage device 424 may be persistent and / or volatile. In some embodiments, storage device 424 may take the form of a quantum computing storage device for storing data in the form of qubits. In embodiments where computer 401 requires a large amount of storage (e.g., when computer 401 locally stores and manages a large database), such storage can be provided by peripheral storage devices designed to store very large amounts of data, such as a storage area network (SAN) shared by multiple geographically distributed computers. The IoT sensor set 425 consists of sensors that can be used for Internet of Things (IoT) applications. For example, one sensor could be a thermometer, and another could be a motion detector.

[0114] Network module 415 is a collection of computer software, hardware, and firmware that allows computer 401 to communicate with other computers via WAN 402. Network module 415 may include hardware such as a modem or Wi-Fi transceiver, software for packetizing and / or depacketizing data transmitted over the communication network, and / or web browser software for transmitting data over the Internet. In some embodiments, the network control and network forwarding functions of network module 415 are performed on the same physical hardware device. In other embodiments (e.g., embodiments utilizing Software-Defined Networking (SDN), the control and forwarding functions of network module 415 are performed on physically separate devices, such that the control function manages multiple different network hardware devices. Computer-readable program instructions for performing the methods of the present invention can typically be downloaded to computer 401 from an external computer or external storage device via a network adapter card or network interface included in network module 415.

[0115] A WAN 402 is any wide area network (e.g., the Internet) capable of transmitting computer data over non-local distances using any existing or future-developed technology for transmitting computer data. In some embodiments, a WAN 402 may be replaced and / or supplemented by a local area network (LAN) designed to transmit data between devices located in a local area, such as a Wi-Fi network. WANs and / or LANs typically include computer hardware such as copper transmission cables, fiber optic transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and edge servers.

[0116] End User Equipment (EUD) 403 is any computer system used and controlled by an end user (e.g., a customer of the enterprise operating computer 401) and can take any form discussed above in connection with computer 401. EUD 403 typically receives useful and helpful data from the operation of computer 401. For example, in a hypothetical scenario where computer 401 is designed to provide recommendations to an end user, these recommendations are typically transmitted from network module 415 of computer 401 to EUD 403 via WAN 402. In this way, EUD 403 can display or otherwise present the recommendations to the end user. In some embodiments, EUD 403 can be a client device, such as a thin client, a heavy client, a mainframe computer, a desktop computer, etc.

[0117] Remote server 404 is any computer system that provides at least some data and / or functionality to computer 401. Remote server 404 can be controlled and used by the same entity operating computer 401. Remote server 404 represents multiple machines that collect and store beneficial and useful data used by other computers, such as computer 401. For example, in a hypothetical scenario where computer 401 is designed and programmed to provide recommendations based on historical data, this historical data can be provided to computer 401 from a remote database 430 of remote server 404.

[0118] Public cloud 405 is any computer system available to multiple entities, providing on-demand availability of computer system resources and / or other computing capabilities (especially data storage (cloud storage) and computing power) without direct active management by the user. Cloud computing typically leverages resource sharing to achieve scalability and cost-effectiveness. Direct and active management of the computing resources of public cloud 405 is performed by the computer hardware and / or software of cloud orchestration module 441. The computing resources provided by public cloud 405 are typically implemented by virtual computing environments running on various computers constituting host physical machine set 442, which is the entire domain of physical computers within public cloud 405 and / or available to public cloud 405. Virtual computing environments (VCEs) typically take the form of virtual machines in virtual machine set 443 and / or containers in container set 444. It should be understood that these VCEs can be stored as images and can be transferred between various physical hosts, either as images or after VCE instantiation. The cloud orchestration module 441 manages the delivery and storage of images, the deployment of new VCE instances, and the management of VCE deployment activity instantiation. Gateway 440 is a collection of computer software, hardware, and firmware that allows the public cloud 405 to communicate via WAN 402.

[0119] Now, we will provide some further explanation of Virtualized Computing Environments (VCEs). A VCE can be stored as an "image." A new active instance of a VCE can be instantiated from this image. Two common types of VCEs are virtual machines and containers. A container is a VCE that uses operating system-level virtualization. This refers to an operating system feature where the kernel allows multiple isolated user-space instances (called containers) to exist. From the perspective of a program running within it, these isolated user-space instances typically behave like a real computer. A computer program running on a regular operating system can utilize all the resources of that computer, such as connected devices, files and folders, network shares, CPU power, and quantifiable hardware capabilities. However, a program running inside a container can only use the contents of that container and the devices allocated to that container; this characteristic is called containerization.

[0120] Private cloud 406 is similar to public cloud 405, except that its computing resources are available only to a single enterprise. While private cloud 406 is depicted communicating with WAN 402, in other embodiments, private cloud can be completely disconnected from the internet and accessible only via a local / private network. A hybrid cloud is a combination of multiple clouds of different types (e.g., private, community, or public cloud types) typically implemented by different vendors. Each of the multiple clouds remains an independent and separate entity, but the larger hybrid cloud architecture is bound together by standardized or proprietary technologies that enable orchestration, management, and / or data / application portability across the multiple component clouds. In this embodiment, both public cloud 405 and private cloud 406 are part of a larger hybrid cloud.

[0121] The various embodiments of the present invention are described for illustrative purposes and are not intended to be exhaustive or limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is intended to best explain the principles of the embodiments, their practical application, or technical improvements to the prior art, or to enable others skilled in the art to understand the embodiments disclosed herein.

[0122] The foregoing can be improved and modified without departing from the scope of the present invention.

Claims

1. A computer-implemented method for assessing the carbon sequestration contribution of nature-based assets located in an aquatic environment, the method comprising: Obtain the meta descriptor of the nature-based asset and the meta descriptor of the area of ​​the aquatic environment where the nature-based asset is located; Map the dimensions of the nature-based assets; Based on the meta-descriptor of the nature-based asset, access data on the monitored physical properties of the nature-based asset in the aquatic environment; Based on the meta-descriptor of the region, access the monitored carbon concentration data in the region of the aquatic environment; Based on the meta-descriptor of the region, access data on the monitored physical, chemical, and biological properties of the aquatic environment; as well as The model is applied to the accessed data to assign a per-unit measured carbon sequestration contribution to the nature-based asset.

2. The method according to claim 1, comprising: The tokenization contributions of the nature-based assets are stored in an immutable data repository.

3. The method according to claim 1 or claim 2, wherein the data on the monitored carbon concentration in the area accessing the aquatic environment includes in-situ data and external environmental data.

4. The method according to any one of claims 1 to 3, comprising: The carbon concentration in the water environment of the area is monitored using in-situ data and remote sensing, and using rapid autonomous monitoring of water variables. as well as Acoustic and / or visual devices are used to monitor the physical characteristics of the nature-based assets in the area.

5. The method according to any one of the preceding claims, comprising applying the model by inputting parameters of the nature-based asset and the meta-descriptor of the region.

6. The method according to any one of the preceding claims, comprising: Location modeling, used to model the locations for deploying sensor devices in the region, includes increasing sampling in areas where the training or fine-tuning of the model contributes more significantly.

7. The method according to any one of the preceding claims, comprising applying the model to process and classify the health index of the nature-based asset, and / or predicting the carbon sequestration of the nature-based asset over time.

8. The method according to any one of the preceding claims, comprising updating data collection requirements based on the performance of the model.

9. The method according to any one of the preceding claims, comprising: The model is trained in the form of an artificial intelligence model to assign the per-unit measured carbon sequestration contribution of the nature-based asset using a combination of the following feature extractions: Carbon concentration estimation and driving factors from aquatic environment data; and Basic asset variables derived from asset monitoring data.

10. The method of claim 9, wherein the training comprises combining the following feature extractions: Learn about the relevant environmental characteristics that affect carbon sequestration potential; Carbon concentration values ​​are estimated based on hydrological and atmospheric environmental variables; Estimate the contribution of different features to the model predictions and the spatial and temporal characteristics of these different features; and Extract information about the physical and material properties of the nature-based assets.

11. The method according to claim 9 or claim 10, wherein the feature Extraction includes point measurements and geospatial measurements of environmental indicators, as well as labels corresponding to nature-based contributions on a high-resolution grid.

12. A system for assessing the carbon sequestration contribution of nature-based assets located in an aquatic environment, comprising: A processor and memory, the memory being configured to provide computer program instructions to the processor to perform the functions of the following components: A meta descriptor component, the meta descriptor component being used to obtain meta descriptors for nature-based assets and meta descriptors for the area of ​​the aquatic environment in which the nature-based assets are located; A mapping component for mapping the size of the nature-based asset; An asset data monitoring component is used to access data on the monitored physical characteristics of the nature-based assets in the aquatic environment based on the meta-descriptor. A carbon concentration data component, the carbon concentration data component being used to access monitored carbon concentration data in the region of the aquatic environment based on the meta-descriptor; A water environment data component, the water environment data component being used to access data on the monitored physical, chemical and biological properties of the water environment based on the meta descriptor of the region; as well as A model application component, which is used to apply a model to assign a per-unit measured carbon sequestration contribution to the nature-based asset.

13. The system of claim 12, comprising: A tokenization component for storing the tokenization contributions of the nature-based asset in an immutable data repository.

14. The system of claim 12 or claim 13, comprising a carbon concentration monitoring system for monitoring carbon concentration in the area of ​​the aquatic environment using in-situ data and remote sensing, and using rapid autonomous monitoring of water variables.

15. The system according to any one of claims 12 to 14, comprising an asset monitoring system for monitoring the physical characteristics of the nature-based assets in the area using acoustic and / or visual devices.

16. The system according to any one of claims 12 to 15, comprising a data collection and update component to update a data collection strategy based on the performance of training the model.

17. The system according to any one of claims 12 to 16, comprising a sensing location modeling system for modeling the location for deploying sensor devices in the region, including increasing sampling in regions that contribute more to training or fine-tuning the model.

18. The system according to any one of claims 12 to 17, comprising communicating with an autonomous surface or underwater vehicle or device to deploy the sensor device at different locations in the region.

19. The system according to any one of claims 12 to 14, comprising a model training system, the model training system being used for: The model is trained in the form of an artificial intelligence model to assign the per-unit measured carbon sequestration contribution of the nature-based asset using a combination of the following feature extractions: Carbon concentration estimation and driving factors from aquatic environment data; and Basic asset variables derived from asset monitoring data.

20. A computer program stored on a computer-readable medium and loadable into the internal memory of a digital computer, the computer program comprising a software code portion configured to perform the steps of the method according to any one of claims 1 to 11 when the computer program is run on a computer.

Images