Method for managing carbon reduction projects

Abstract

A method for classifying an environmental project implemented by a platform 10 for managing environmental projects comprises obtaining project data associated with an emission avoidance project or an

Description

Field The present application relates to a method and a system for managing carbon reduction projects. Background Carbon reduction projects absorb, reduce or prevent emissions of CO2 or other greenhouse gases. These projects are eligible to generate carbon credits, which are exchanged through the Voluntary Carbon Market, VCM. One carbon credit corresponds to one tonne of CO2 absorbed or reduced by the projects. Today, the market price of a carbon credit reflects the CO2 reduction capacity of the project. Producers of greenhouse gases can choose to purchase carbon credits through the Voluntary Carbon Market to compensate for their CO2 emissions and achieve carbon offsetting. Carbon offsetting or climate contributions is a practice adopted by organisations to improve their sustainability strategies and contribute toward achieving net-zero emissions. Today a relatively small number of types of emission reduction projects generating carbon credits are available or in existence. The following types of projects are available today. Forestry and land use projects such as REDD+ (Reducing Emissions from Deforestation and forest Degradation) are projects that preserve and restore existing forest areas threatened by deforestation. Emission reduction projects include those harnessing renewable power infrastructures and sustainable energy sources providing alternatives to the fossil fuels. These projects produce renewable power and contribute to the decarbonization of traditional energy systems. Certain projects focus on providing energy efficient home appliances and stoves that rely on biogas or other sustainable fuel in regions where burning wood is much used to produce energy. Other known projects include the supply of clean water in rural areas, which removes the need to boil water and reduce Greenhouse Gas, GHG, emissions. Certain projects also protect and restore natural ecosystems, such as mangroves, while others capture the methane released into the atmosphere to turn it into fuel. Agriculture is a breeding ground for emission reduction projects since several practices have been established to use agricultural lands and products to store carbon into soils while restoring biodiversity. All these projects create opportunities to capture and / or avoid GHG emissions to reach global carbon neutrality, while earning additional income. Aspects and embodiments were conceived with the foregoing in mind. Summary There is provided a method that facilitates capturing details of a carbon capture project. The method determines and makes available a class or a characteristic of the project in a secure and reliable manner. In an aspect, there is provided a computer implemented method for classifying an environmental project. In one example, an environmental project may comprise one or more of a carbon capture project, an emission reduction project or an emission avoidance project. In other examples, an environmental project may be any scheme which is configured to remove or absorb carbon emissions from the surrounding environment. The method may be implemented by one or more processors or processing resources associated with a platform for managing environmental projects. The processors or processing resources may be hardware or software implemented or may be cloud implemented. The method may comprise the steps of: obtaining project data associated with a project. The project data may be obtained via a user interface provided on a computing device. The project may be an emission avoidance project (e.g. a scheme which prevents carbon being released into the atmosphere) or an emission reduction project (e.g a scheme which removes carbon from the atmosphere or reduces the amount of carbon emitted into the atmosphere). The method may further comprise determining a characteristic of the project based on one or more attributes of the project data that relates to a feature of interest in a data set associated with a data model trained for classifying environmental projects. A data model may be implemented using one or more trained models. A trained model may be implemented using the application of an artificial neural network (ANN) or a convolutional neural network (CNN), which is trained as set out above. ANNs can be hardware - (neurons are represented by physical components) or software-based (computer models) and can use a variety of topologies and learning algorithms. ANNs usually have at least three layers that are interconnected. The first layer consists of input neurons. Those neurons send data on to the second layer, referred to a hidden layer which implements a function and which in turn sends the output neurons to the third layer. There may be a plurality of hidden layers in the ANN. With respect to the number of neurons in the input iayer, this parameter is based on training data. The second or hidden layer in a neural network implements one or more functions. For example, the function or functions may each compute a linear transformation or a classification of the previous layer or compute logical functions. For instance, considering that the input vector can be represented as x, the hidden layer functions as h and the output as y, then the ANN may be understood as implementing a function fusing the second or hidden iayer that maps from x to h and another function g that maps from h to y. So the hidden layer’s activation is f(x) and the output of the network is g(f(x)). CNNs can also be hardware or software based and can also use a variety of topologies and learning algorithms. A CNN usually comprises at least one convolutional layer where a feature map is generated by the application of a kernel matrix to an input image. This is followed by at least one pooling layer and a fully connected layer, which deploys a multilayer perceptron which comprises at least an input layer, at least one hidden layer and an output layer. The at least one hidden layer applies weights to the output of the pooling layer to determine an output prediction. The training of the trained model may be implemented using any suitable approach such as, for example, supervised, semi-supervised or unsupervised learning. The trained model may alternatively or additionally deploy one or more large language models which are trained on contexts associated with one or more of environmental projects, emission avoidance, emission reduction or carbon capture. The determined characteristic may be indicative of a capacity of the project to avoid emission of or remove a greenhouse gas, GHG, from the air; The method may further comprise assigning the determined characteristic to the project. This may be, by, for example, storing the determined characteristic in a data set which may be associated with one or more tokens which are suitable for incorporation into a blockchain transaction. Optionally, the method further comprising, training the data model to classify environmental projects including the steps of: providing a plurality of data sets, each pertaining to a type of environmental project; for a given data set, providing a training input including a label for a feature of interest associated with the given data set; and providing a training output identifying a characteristic that is associated with the feature of interest pertaining to the label. Optionally, the method further including: providing input project data to the trained data model; the method further comprises, implementing using the trained data model the steps of: detecting at least one feature in the input project data that corresponds to a feature of interest in the plurality of data sets; determining that the at least one feature in the input project data is associated with a label in the training input that corresponds to said feature of interest; identifying that the input project data is of the characteristic associated with said label; and providing as an output the type of environmental project based on the identified characteristic. Optionally, based on one or more attributes of the project data, selecting a trained data model amongst a plurality of data models trained for classifying environmental projects; and determining the characteristic of the project by providing the project data to the selected trained data model. Optionally, obtaining project data comprises obtaining a first data; and based on the selected trained model, obtaining a second data from a trusted source of information; and determining the characteristic of the project by providing the first data, and the second data to the selected trained data model. Optionally, obtaining project data comprises verifying the project data by means of a validation module; and determining the characteristic of the project by providing the verified data to the trained data model. Optionally, the project has an expected Lifecycle, and wherein the project data comprises a period of the expected Lifecycle of the location. Optionally, obtaining project data associated with the project comprises: deriving further data associated with the project to provide enhanced data; and determining the characteristic of the project based on the data including the enhanced data. Optionally, obtaining project data comprises deriving a location data of the project using a location identification technique such as GPS location tracking technique. Optionally, the location data comprises a data set defining a geographic space surrounding the location. Optionally, the location data comprises at least a geospatial data set. Optionally, the location data is captured via a digital map which shows the physical location of the project. Optionally, the data associated with the project comprises a metadata associated with a geographical location pertaining to said project; the method including capturing the metadata associated with the project from a digital image of the location. Optionally, the project is associated with a biological element capable of capturing and storing GHG, so as to absorb or to sequester GHG from the air, the biological element comprising at least one of a plant, an agricultural element, a tree, a living organism such as a microbe or a bacteria, a fungus, an insect such as a bug. Optionally, the project is associated with a mechanical element capable of capturing and storing GHG, so as to absorb or to sequester GHG from the air, wherein the mechanical element comprises a material including at least one of graphene, concrete mixtures containing GHG absorbing ingredients. Optionally, the project is associated with at least one of a piece of land, an amount of a mineral, a liquid such as an aqueous solution capable of removing GHG from the air. Optionally, determining the characteristic of the project comprises determining a volume or a mass of GHG expected to be captured by a location associated with the project during a period of an expected Lifecycle of the project. Optionally, the GHG is carbon dioxide, the method further comprising using the determined characteristic of the project for issuing or updating a rating to a carbon credit associated with the project; or issuing or updating a carbon credit associated with the project. Optionally, the method comprises storing the determined characteristic as metadata associated with the project, said metadata being stored on a blockchain or on an off-block storage resource. Optionally, the the project is represented by a digital token for a given geographical location, and the digital token is associated with a Lifetime of the project; wherein the token expires upon expiry of the Lifetime. Optionally, the token is associated with an output of a blockchain transaction. Optionally, the method comprises determining whether the project is associated with an existing token; wherein if the project is associated with an existing token; linking the existing token to the determined characteristic and wherein if the project is not associated with an existing token; generating a token for the project based on the project data. Optionally, generating the token comprises extracting a geographical location and an identity of an entity associated with the project from the project data; allocating the extracted location, the identity and determined characteristic associated with the project to the token. In an aspect, there is provided a computer implemented method of registering a project associated with a plurality of locations, the method comprising: capturing data associated with the plurality of locations; allocating a project token to each location among the plurality and to the identity of an entity associated with project and / or given location; providing the captured data into a trained model; determining, using the trained model, a characteristic of the project based on the captured data associated with the plurality of location; wherein the characteristic of the project is indicative of a capacity of the project to avoid or remove a greenhouse gas, GHG, from the atmosphere; assigning the characteristic of the project to the allocated project token. In an aspect, there is provided a computer implemented method of registering a location. The location is an area that comprises one or more items capable of avoiding emission of or remove a greenhouse gas, GHG, from the air. The method comprising obtaining data 5 associated with a token, the token corresponding to a location, wherein the token is suitable for incorporation into a blockchain transaction; providing the obtained data into a trained model; obtaining , using the trained model, a characteristic of the location based on the obtained data; wherein the characteristic of the location is indicative of a capacity of the location to avoid emission of or remove a greenhouse gas, GHG, from the air; assigning the 10 characteristic of the location to the token. Brief Description of the Drawings Embodiments are now described by way of example only and with reference to the following drawings in which: Figure 1 shows an exemplified architecture of a system comprising a software platform in accordance with embodiments of the present disclosure. Figure 2 shows a flow diagram illustrating an exemplary method for managing carbon reduction projects in accordance with embodiments of the present disclosure. Figure 3 shows a flow diagram illustrating an exemplary method for determining a characteristic of a project using a trained model based on data captured in accordance with embodiments of the present disclosure. Figure 4 shows an example of classification in accordance with embodiments of the present disclosure. Figure 5 illustrates a stack showing different layers of a system that enables managing carbon reduction projects. Detailed description Today, there is a range of different types of emissions reduction projects participating in the Voluntary Carbon Market, VCM, around the globe, and this range is expected to grow further. Organizations including the World Bank, the International Organization of Securities Commissions, IOSCO, and the Integrity Council for the Voluntary Carbon Market, ICVCM, have been developing a set of standards for the VCM. These standards define a framework to set voluntary global threshold standards for quality in the voluntary carbon market. This framework is to help buyers recognize and put a price on carbon credits. The price of carbon credits depends on the quality and the integrity of the project that originate them. However, there is no taxonomy available today as there is no categorization or classification of these projects that would help market participants recognize and identify high-integrity, high quality projects. The present application enables the implementation of secure and immutable technical means of achieving a classification of these projects into groups or types. Achieving such categorization will help the public to identify real characteristics of carbon absorbing projects around the world and will also enable the impact of these projects to be monitored and audited. These characteristics when provided as part of a common taxonomy will help the VCM participants (purchasers / sellers of carbon credits) to correctly assess the value of different projects. This in turn facilitates pricing carbon credits issued for these projects. The 8 classification helps organisations select projects that are high-quality emission reductions projects, and which meet their needs and objectives. The classification obtained comprises attributes which can be standardised and measured. Various types of carbon capture projects include biological and mechanical projects and various possible sub-categories. To this end, and referring to FIGURE 1, the present application provides a software platform 10 which enables location data associated with a location and assets at that location to be securely and immutably associated with carbon capture characteristics. The software platform 10 comprises a plurality of modules and components which will now be described. The modules and components may be hardware or software implemented. The modules and components may be co-located at a single location or distributed over several locations. The modules and components are configured to communicate using any suitable data communications means or protocol. The software platform 10 enables entities to register carbon reduction projects. Details of the projects can be entered into the software platform via a capture module 103. The software platform 10 is provided with an interface to a blockchain so that the data captured through the platform can be used to create and execute smart contracts recorded onto the blockchain. The software platform 10 is provided with a Classification Engine 109 which analyses the data information captured. The classification engine 109 classifies the project whose data is captured by producing an output which characterises the carbon reduction project. The Classification Engine 109 comprises applying a machine learning model and an algorithm to the received data to produce a characteristic of the project. The characteristic of the project comprises a classification which accurately identifies a capacity of the project to avoid emission or remove greenhouse gas from the atmosphere. The classification may comprise an alphanumeric identifier and / or a number. The classification may be sized (in terms of storage requirement) in a way which enables it to be included in a blockchain transaction as metadata associated with an output of that transaction. The classification engine or classifier, is based on machine learning involving a machine learning algorithm and categorizes or classifies input data into predefined classes or categories. The classification engine produces a ‘Class’ or an Identifier of a classification assigned to a project according to its characteristics. The classification engine takes a proforma input (a metadata file based on tags fields) and auto-create a Voluntary Carbon Credit Contract Specification. The tags identify specific parameters relating to the project and / or the location of the project. The algorithm of the classification engine is trained on a labelled dataset, such that each data point is associated with its corresponding class label. Once trained, the classification engine can be used to predict the class label of new, unseen data based on the patterns it has learned during training. A goal of the classification engine is to map input data to the most suitable class based on the features present in the data. The classification engine uses statistical methods and mathematical models to identify patterns and relationships between features and classes in the training data. Commonly used algorithms for classification tasks include Decision Trees, Random Forests, Support Vector Machines (SVM), Logistic Regression, and Neural Networks. A data hierarchy within the Biological and Mechanical project sectors can assist in streamlining the process of the classification engine. Table of FIGURE 4 exemplary shows a hierarchy within Biological and mechanical projects. Other examples of classifications not visible in the exemplary Table of FIGURE 4 are within the scope of the present disclosure. For instance, a hierarchical classification can include at least the following sectors: “Energy Industries”, “Transport”, “Carbon Capture and Storage”, “Forest Conservation and Reforestation Projects”. The “Energy Industries” sector can include at least the following sub-sectors: Renewable Energy Projects, Energy Efficiency Projects, Energy distribution, Energy demand. The Renewable Energy Projects sub-sector can further include Solar power installations, and at least wind farms. Wind farms can be subdivided into Wind Energy, Wind Turbines, Wind Power Generation, Wind Resource Assessment, Turbine Efficiency. The Energy Efficient Projects sub-sector can include at least Hydroelectric power plants, Geothermal energy projects, Biomass energy projects, Biogas, Building retrofits for improved energy efficiency, Energy-efficient lighting systems, Energy management systems. The “Transport” sector can include at least Electrification of transportation fleets, Public transit and mass transportation systems, Carpooling and ridesharing initiatives, Active transportation, Alternative fuels, Electric vehicles (EVs) etc. The “Carbon Capture and Storage” sector may include at least Post-combustion capture in power plants, Pre-combustion capture in fossil fuel processing, Direct Air Capture (DAC) systems, Industrial process optimization, Carbon storage and carbon removal projects, etc. The “Forest Conservation and Reforestation Projects” sector may comprise at least Avoided deforestation initiatives, Forest conservation programs, Afforestation and reforestation projects, Tree planting, Tree nurseries etc. The following provides exemplary steps that can be carried out to design and build a classification engine. These steps are described a follows: 1. Data Collection: Accumulate a diverse set of metadata corresponding to the sectors listed in the TABLE about different projects associated with carbon credits. This should be collected from various sources of metadata, such as industry reports, research publications, or project databases. These files should contain enough detailed information to allow the classification engine to distinguish between the various categories. Afterwards, the format of the collected data, e.g., CSV, JSON, or XML can be standardized. This will make up a tagged ontology library. An ontology is a formal representation of knowledge in a specific domain, consisting of concepts, attributes, and relationships between entities. A tagged ontology library is a collection of documents or data points with relevant tags or class labels already assigned, which can be used as the labelled training dataset for the classification engine. 2. Data Preprocessing: Clean and pre-process the data to ensure it is uniform and free of any inconsistencies, duplicates and errors. This might involve converting all text to lowercase, removing special characters, or filling in missing values. 3. Tagged Ontology Library Ingestion: Ingest or input the tagged ontology library, which contains pre-classified data points relevant to the voluntary carbon credit market. Each data point should be associated with a specific class label or category. 4. Feature Extraction: Identify the key features or tags in the metadata files that would be instrumental in classifying a given file into one of the sectors listed. These features could include project attributes, carbon offset methodologies, emission reduction levels, geographical locations, project types, and other relevant data. 5. Splitting the Dataset: Split the dataset into training and testing sets. The training set will be used to train the classification engine, while the testing set will be used to evaluate its performance. 6. Training the Classification Engine: Use the training set to train the classification engine using appropriate machine learning algorithms. The engine will learn the patterns and relationships between the features and class labels from the tagged ontology library. 7. Evaluation and Fine-tuning: After training, assess the performance of the classification engine on the testing set. Fine-tune the model and hyperparameters to optimize its accuracy, precision, recall, and score. 8. Deployment and Integration: Deploy the trained classification engine in a production environment where it can process new, unseen data and predict the appropriate class labels based on the learned patterns. Integrate the trained classification engine with the system that would auto-create a Voluntary Carbon Credit Contract Specification. For example: • API Creation: Convert the model into an API for easy integration. • Scalability: Ensure the model can handle large volumes of requests, possibly by deploying it in a cloud environment. 9. Feedback Loop: Implement a feedback mechanism. Users can provide feedback on the accuracy of the classifications, and this feedback can be used to further refine and train the model. 10. Maintenance: • Monitoring: Continuously monitor the system for any discrepancies or issues. • Updates: As new sectors emerge or existing ones evolve, update the model and hierarchy accordingly. By following this hierarchy and the steps outlined a robust classification engine could be designed that efficiently categorizes metadata files and aids in the auto-creation of Voluntary Carbon Credit Contract Specifications. When developing a classification engine for the voluntary carbon credit market, several factors should be considered to ensure its accuracy and relevance: 1. Project Attributes: Consider relevant project attributes, such as project type (e.g., afforestation, renewable energy), location, project size, and technology used. 2. Carbon Offset Methodologies: Incorporate information about different carbon offset methodologies, such as avoided deforestation, soil carbon sequestration, renewable energy generation, etc. 3. Emission Reduction Levels: Account for the level of emission reductions achieved by each project and its impact on climate change mitigation. 4. Project Credibility: Assess the credibility and verification status of carbon credit projects, including third-party validation and verification. 5. Sustainable Development Co-benefits: Consider the positive social and environmental co-benefits associated with each project, such as community engagement, biodiversity protection, and sustainable development contributions. 6. Project Lifecycle: Take into account the stages of each carbon credit project, from planning to implementation and monitoring. 7. Voluntary vs. Compliance Market: Differentiate between projects aiming to generate voluntary carbon credits and those participating in compliance markets or regulated schemes. 8. Geographical Context: Consider regional or country-specific factors, as the carbon credit market can vary significantly based on regional policies, regulations, and market dynamics. By integrating these factors into the classification engine's training data and feature extraction process, the model can make informed and accurate predictions regarding carbon credit projects' classifications and contributions to climate action and sustainable development. Additionally, regular updates and maintenance of the model are essential to keep it relevant and aligned with evolving market dynamics and practices. In some examples, the software platform 10 utilises a Token that uniquely identifies the project. There is provided a Token which is allocated to a physical location to give the public certainty that the project is real and legitimate. FIGURE 1 illustrates a module 113 which issues the Token. The module may reside within the platform 10 or be external to the platform 10. The Token may also be associated or allocated to a global Identifier of an entity that is associated with a carbon capture project. Therefore, the Token is a geospatial Token that uniquely Identifies a legal person associated with the project such as the owner of the project. The geospatial Token therefore uniquely identifies a carbon capture project located at a given location and being associated with a legal person. Consequently, the geospatial Token is used to uniquely identify carbon credits of the corresponding project. A geospatial Token associated with a project issuing carbon credits advantageously prevents recording the same Token in respect of two different projects taking place at two different locations. Therefore, the geospatial Token prevents frauds that occur when a token of a carbon credit issued by a given carbon credit project is offered on the market more than once as representing two or more different projects. Moreover, the blockchain technology ensures immutability of the recording. Thus, the blockchain prevents that a geospatial Token representing a project at a given location is tampered with to represent a different project at a different location. The infrastructure using blockchain and the geospatial Token offers robust protection against the multiple use of a single Token for different projects at different locations. It prevents frauds that occur today in the VCM whereby entities issue carbon credits that they sell multiple times to several buyers by making them available numerous times on different registers. There is provided a method of registering a carbon capture project which operates at a specific location to sequester GHG (such as, for example, CO2) from the atmosphere. The software platform 10 according to examples is provided with a user interface for capturing details of a given project. The data relating to a specific project may be captured via a web portal for instance. Referring to FIGURE 2, the registration of a project comprises an on-boarding process 201 through which a user can enter the details of their projects. The user may represent an entity associated with the project and may, for example, be a person acting on behalf of the owner of the project. The data can be captured via the on-boarding process during which a user is guided for entering relevant data concerning the project. The data may be entered on the platform manually by the user and / or by a process which retrieves the data directly from the relevant sources of information. As indicated, the software platform comprises a capture module 103, which stores any data entered and received on the platform or may be configured to interface with local or remote storage to store any data entered and received on the platform. The data entered on the platform includes location data associated with the physical location of the project. The location data of the location of the project comprises a data set defining a geographic space surrounding the project. The data set may also define a plurality of individual items within the project. For instance, where a given project comprises a collection of trees, each tree may be identified by its own location data on the platform. The location data includes for instance a geospatial data, such as for instance GPS information, including longitude, latitude, and altitude. The location may be identified using a What3Words ® address. The geospatial data may correspond to coordinates identifying a unique point or a set of coordinates whereby each set of coordinates uniquely identifies a particular item. The data entered on the platform also includes a legal identity that uniquely identifies an entity associated with the project (e.g ID of the owner / licensee of the project). In a system comprising the software platform 10, there is provided a trusted independent third party 107 and a secured interface between the software platform and the trusted party. Data entered on the platform 10 through the user interface are verified and validated through the secured interface. In an implementation, the software platform 10 is provided with a validation module 105 which intercepts the data entered through the user interface (e.g the web portal) and verifies the data by cooperating with the independent third party 107 over the secure interface and / or other trusted sources 111. Referring to FIGURE 2, the registration of a project is accompanied by the step of verification 203 of the data entered through the onboarding process 201. Verification of data is intended to ensure that the data entered including the legal identity of the entity associated with the project are real, trustworthy and legitimate. The verification may employ one or more suitable verification and / or authentication protocols to validate different types of data. The verification of the legal identity of the entity associated with the project may be carried out by a suitable ID verification method. Suitable ID verification methods include for instance methods that use data gathered from a variety of trusted sources 111. Online verification techniques that employ Artificial Intelligence, Al, biometrics, computer vision may also be used to determine for instance if a government ID exists for the entity associated with the project. The data entered on the platform may include information regarding an amount of CO2 sequestered by a given project which is being registered. These data can be detailed to inform the platform about how much CO2 had been sequestered by the project in a period such as since its creation. The data captured on the platform may also include an age of the project and an expected lifetime of the project since its creation. This may include the expected remaining lifetime of the project since its registration on the platform. The capture module 103 stores the data entered on the platform 10 and this may include the data indicating a quantity of CO2 that had been sequestered by the project and / or the quantity of CO2 expected to be sequestered by the project for the remainder of its lifetime. The software platform 10 also verifies through the validation module 105 the data captured detailing the carbon capture project being registered. The validation module 105 may verify the data indicating the quantity of CO2 sequestered and / or expected to be sequestered by the project for the remainder of its lifetime. There is provided a validation engine for instance running as part of the validation module 105, which performs an independent assessment of the volume or quantity of CO2 absorbed by the project. To undertake a measurement of the volume or quantity of CO2 absorbed by the project, the validation module 105 may use certain existing models. For instance, for a project consisting of a forest and trees, the validation module 105 may use models that estimate forest biomass accurately. The forest biomass gives an indication of the quantity of CO2 absorbed by a tree or a forest and therefore gives an indication of the absorption capacity of the project. Referring to FIGURE 2, there is provided a step 205 of injecting the verified data into the classification engine. As indicated above, the Classification Engine 109 analyses the data information captured. It classifies the project whose data is captured by producing an output which characterises the carbon reduction project. A further detailed example of the functioning of the Classification Engine 109 is provided below with reference to FIGURE 3. Referring to Figure 2, step 209 identifies an output produced by the Classification Engine 109. According to this disclosure, the output is obtained using a trained model of the Classification Engine 109. In an example, the output is a ‘Class’ or an Identifier of a classification assigned to the project according to its characteristics. The ‘Class’ can be used to appreciate a quality of the carbon capture project. Step 211 is a test to determine whether the location of a project is associated with an existing Token. As indicated above, the data entered on the software platform 10 includes location data associated with the physical location of the project. If the project is associated with an existing Token, the test step 211 of the process of FIGURE 2 leads to step 213 via the branch indicated by ‘yes’. At step 213, the existing token is linked to the project and to a legal identity that uniquely identifies an entity associated with the project. If the project is not associated with an existing Token, the test step 211 leads to step 215 via the branch indicated by ‘no’, at step 215, a new Token is generated for the project and for the legal identity that uniquely identifies the entity associated with the project. Alternatively, a pregenerated Token is allocated to the project. A pre-generated Token is reserved for a given geospatial location until it is claimed for a project or allocated to a project, so that once allocated to a project at a given location, the token is not available anymore for a different project at a different location. Throughout this specification, the terms Token and geospatial Token are used interchangeably. According to this disclosure, several ways can be devised to generate a geospatial Token which uniquely identifies an existing physical location. Generating the geospatial Token comprises generating the Token based on an existing location (the location of the project) and on the identity of the entity associated with the location (e.g ID of the owner of the project). In an example, the Token is generated based on the location of the project and on the identity of the entity associated with the location, which have been captured through the software platform. The data captured includes location data associated with the physical location of the project, and an identity such as the ID of the owner of the project. In a further example, generating the Token involves computing a hash function of a location data representative of the location of the project. In an example, generating the Token involves computing a hash function of a combination of a location data representative of the location of the project and the identity of the entity associated with the location. In this way the Token in the form of a Hash is only obtained by the correct location of the project or the correct combination of the location of the project and the identity of the project owner. A geospatial Token generated in this manner provides enhanced security as when it is allocated to a project at an existing location it cannot be used for a different project at a different location. It is noted that the order of the steps of FIGURE 2 can be altered in various examples. For instance the test step 211 can be performed during or just after the on-boarding process of step 201 such that an allocation of a Token to a project can occur before , simultaneously with or contemporaneously with step 205 of injecting captured data into the Classification Engine. Referring to step 217 of FIGURE 2, the Token associated with the project is stored together with the characteristic of the project obtained by the trained model on a blockchain 115. In some examples, the Token may be associated with an unspent transaction output (UTXO) or with a provably unspendable output through use of the OP_RETURN opcode. Advantageously the Bitcoin Satoshi Vision, BSV, blockchain is used as the underlying blockchain 115. As previously indicated, the Token serves to uniquely identify a project or an asset deployed at the project, as being located at the location for which the token is allocated. The Token is therefore used to uniquely identify the characteristic of the project. The storage of the Token together with the characteristic(s) of the project on a blockchain 115 is provided with the security benefits of the blockchain including immutability and integrity of the recording. Hence the public has confidence that the Token and the characteristic of the corresponding project as recorded on the blockchain have not been altered. The integrity of the recording of the Token and the characteristic of the corresponding project help to ensure that the carbon reduction project and its characteristic are real, legitimate and trustworthy. The system including the platform and the blockchain also helps to provide proof or origin of the projects. Since, a carbon credit is typically issued fora carbon reduction project, the protection provided to this recording helps ensure that the carbon credit of the project is legitimate and that it represents a genuine quantity of absorption capacity of the project. In turn this information ensures that the carbon credit of a project registered through the platform achieves a market price that reliably reflects the quality of the project in terms of CO2 absorption capacity. Reliability of data associated with a carbon credit helps the market to quickly discover the fair price of the carbon credit corresponding to a carbon capture / reduction project. Step 219 exemplary shows that the Tokens visible on a blockchain facilitate market price discovery of the carbon credits of the corresponding projects on the VCM. The Bitcoin Satoshi Vision, SV, blockchain or BSV blockchain is selected for its scalability, and efficiency as the underlying blockchain to store information relating to carbon capture projects. The scalability of the BSV blockchain is derived from its block sizes. Owing to its large block sizes, the Bitcoin SV network can support over 50,000 transactions per second, which enables the BSV to finalize or settle transactions almost instantaneously and makes the Bitcoin SV network more scalable than that of Bitcoin. This contrasts with Bitcoin whose peak transaction throughput is only around 7 transactions per second. In addition to the abovementioned advantages, the BSV blockchain is cost-effective as its transaction fees are significantly lower than that of the Bitcoin blockchain and requires lower energy for mining. The scope of this disclosure is not however limited to use of a specific ledger and, it will be appreciated that some different versions may implement other types of blockchains. In a particular implementation the software platform has several interfaces to different blockchains for recording the Tokens together with the characteristic(s) of a project on multiple blockchains. In a further implementation, the software platform is configured to select a target blockchain for storage of the characteristics and / or data according to the characteristics and / or data relating to the project. For instance, characteristics and / or data of different types of carbon capture projects may be recorded on different types of blockchains. Various types of carbon capture projects include biological and mechanical projects and various possible subcategories. FIGURE 4 gives an example of classification of carbon capture projects in which the last row shows arbitrary classes (Class of Type of project) as generated by the classification engine. As the technology evolves and new types of blockchains are created, certain blockchains may appear more suitable for different types of carbon capture projects. For instance, certain blockchains may be recognised as suitable for biological projects whereas others may be more suitable for mechanical projects etc. The information relating to a carbon capture project including the characteristics (class as obtained by the classification engine), the Tokens, and / or the data obtained through the platform may be used to create and execute smart contracts. Smart contracts are typically executed to buy and sell assets. For instance, smart contracts can be executed to trade carbon credits and / or Tokens identifying carbon capture projects which issue carbon credits. Step 221 illustrates that the geospatial Tokens according to this disclosure are traded via the execution of smart contracts. The smart contracts can be stored on the blockchain that stores the data relating to the carbon capture project or on a different one. The smart contracts are created according to standardised set of rules. Several standards are typically used to define the terms to be applied to a smart contract between parties. For instance, the International Swaps and Derivatives Association or ISDA Master Agreement is used to define the terms to be applied to smart contracts to effect over the counter derivatives transactions between parties. The ISDA Master Agreement is used for trading various asset classes including bonds, interest rates instruments, commodities, and more recently digital assets such as carbon credits. The blockchain is then used to provide all the security benefits to the information (e.g smart contract) recorded to the blockchain such as the verification of the integrity of the information. Transactions can be settled in an underlying cryptocurrency of the blockchain such as the BSV. Blockchain such as BSV blockchain is used as a settlement layer. The characteristics (classes) of different projects are indicative of a capacity of the respective project to avoid emission of or remove a greenhouse gas, GHG, such as CO2 from the air. Therefore, for each project registered, the association of the Token and the obtained characteristic, provides an indication of the value of the Token or of the corresponding carbon credit(s). Since the Blockchain and its contents are publicly available, rating agencies active in the VCM such as BeZero can consult the Tokens, and Characteristics (classes) of various carbon capture projects. They can use the Characteristics (classes) of various carbon capture projects to complement their own assessment based on fields measurements to attribute ratings to these tokens / carbon credits. Step 223 exemplary illustrates how rating agencies can draw on the available information obtained through the software platform to give a rating to different Tokens or carbon credits of corresponding projects. The software platform provides an enhancement or enrichment layer as the information it produces including the indication of a capacity of the project to avoid or remove CO2 from the air, can influence an initial market price of a Token / carbon credit. The software platform provides a means to verify that projects are real, legitimate and authorised to issue carbon credit. It also provides a means to verify genuine quantity of available carbon credits and can help companies manage offsetting their carbon emissions and meet their carbon emission targets. FIGURE 3 illustrates a flow diagram of a method of determining a characteristic of a carbon capture project using a trained model based on data captured through the software platform 10. Referring to step 301, the capture module 103 stores data entered through a user interface such as a web portal. The data relating to a carbon capture project received through an onboarding process for instance to register the project contains various types of data. Certain data associated with the project comprises a metadata. The capture module 103 extracts the metadata from the data captured and stores the metadata. Accordingly, capturing the data associated with the project comprises capturing the metadata associated with the project. In an example where the data captured includes one or digital images of the project uploaded onto the software platform, the capture module 103 extracts the metadata from the digital image(s). The metadata may include one or more tags identifying specific parameters relating to the project and / or the location of the project. In an example, capturing the data by the capture module 103 comprises deriving location data of the location of the project based on the metadata associated with the project. Step 301 exemplary shows that the capture module 103 receives data associated with a project through a user interface for instance and derives further enhanced data based on the received data. To this end, the capture module 103 requests from other sources 111 complementary data to enhance the set of data already captured. Data captured by the capture module 103 and validated by the validation module 105 is entered into the Classification Engine 109. The data injected in the classification engine 109 may include data obtained directly via a user interface during the registration of the project as well as data derived from the data captured via the user interface. The Classification Engine 109 computes a characteristic of each project being registered. The characteristic of a project indicates a capacity of the project to avoid emission of or remove a greenhouse gas, GHG, such as CO2 from the atmosphere. To measure as accurately as possible a capacity of the project to avoid emission of or remove a greenhouse gas, the classification engine 109 classifies the project into a type and applies a CO2 volume calculation model specific to the type of the project. According to the data receives into the Classification Engine, different projects may be identified as belonging to different types. Step 303 of FIGURE 3 illustrates the identification of a project type by the classification engine 109. The Classification Engine may classify the different projects as being one of a biological or a mechanical project. It further identifies a type of project being one of forestry and land, agriculture, farming, capture methane gas and transform to a solid form, supply clean water, re-forestation, renewable power etc. Several sub-types may be arbitrarily defined. The table of FIGURE 4 shows an example of classification comprising several types of projects. The classification Engine 109 obtains if needs be further data (further enhanced data) from other sources 111 in addition to the data captured by the capture module via the user interface to obtain a more detailed description of the project. Step 301 includes the capture of enhanced data further captured by the classification engine 109. Based on the data captured directly via a user interface and / or derived from other sources 111, the classification engine identifies a project type for the project being registered and the corresponding model specific to the type of project identified. A CO2 volume calculation model specific to the type of project is identified at step 305. At step 307, a test is performed as to whether the data available for the project is sufficient to run the specific model. In the event of the affirmative, the method applies the data relating to the project to the model specific to the type of the project to measure its CO2 absorption capacity (step 309). In the event that data captured and / or already derived is not enough for the identified model, the classification engine requests additional relevant data from various trusted sources 111 (step 311). The classification engine finally applies the collection of data including data directly captured and any enhanced data as well as any additional data obtained from various sources 111 to the specific identified model (step 313). As a result of applying the data relating to the project to the identified model, the classification engine computes a GHG absorption capacity of the project (step 315). The GHG absorption capacity of the project may refer to the volume or the mass of CO2 that the project is capable to remove from the atmosphere per year or for the rest of its expected lifetime. The Classification engine also outputs a Class of the project, which is an Identifier of a classification assigned to the project according to its characteristics (step 317). The Class can be used to appreciate a quality of the project. The result or output of the classification engine is recorded to the blockchain 115. In an implementation, the results of the classification engine as well as the data captured relating to the carbon reduction project are recorded to the blockchain. In a further implementation, the output of the classification engine as well as the data captured, and the input data injected into the classification engine are recorded to the blockchain 115. An example of applying data relating to a project to a trained model is provided below for illustration purposes only. If a project is identified to be a forest comprising multiple trees, the model that calculates the volume / mass of CO2 absorbed is for instance one of the models that quantifies the amount of CO2 sequestered by the forest. Such a model estimates forest biomass accurately for instance. For these estimation models, the biomass weight is relating to measurements of tree sizes that can be taken, such as diameter and height. Tree species, and wood density of the tree, which is dependent on the tree species, are also important factors in the calculation of the biomass. A tree having a higher wood density tends to have more biomass than a tree having a lower wood density. Models that estimate biomass based on physical attributes are called allometric equations. These provide estimates of the aboveground biomass of a tree as well as a tree’s belowground biomass. The total biomass for a given tree is calculated as the sum of the aboveground biomass and the belowground biomass. A tree’s total biomass gives an indication of the amount of CO2 captured by the tree. The biomass of a forest or a project comprising multiple trees, can be obtained by a sum of biomasses of the individual trees. The biomass of a tree or a forest gives an indication of the quantity of CO2 absorbed by the tree or the forest in the past and therefore gives an indication of the absorption capacity of the project. To measure exactly how much CO2 a project such as a forest sequesters, the classification engine, obtains additional data to complete a detailed description of the project from various sources 111. Obtaining additional data to describe the project includes based on the location of the project, identifying or defining a geographic space surrounding the project. The exact contours of the area covered by the project being registered can be obtained from a database such as a digital map and / or a set of satellite data. These can be gathered by satellite and electronic imaging systems. Given the identified area of the project, the Classification engine 109, further collects detailed data relating to the individual items (e.g trees) of the project to obtain an accurate picture. Detailed data of the project and its individual items can be obtained by a technology such as Light Detection and Ranging or Lidar, which utilises high-resolution remote sensing. Using such a high-resolution remote sensing or other imaging techniques provides information about the detailed structure of the forest. For instance, the classification engine can further obtain details of each tree within the forest including a species of the tree, an age of the tree, a weight of the tree, a height of the tree, a diameter of the trunk of the tree etc. Other information relating to the whole project can also be obtained including the number of trees, a density of the trees within the forest. Other information relevant to the whole project includes how uniform is the structure of the forest, whether trees are similar and / or of the similar age. By means of a remote sensing or other imaging technique specific vegetation in scattered plots devoid of trees can be identified as well as different types of growth areas. Different plots of the forest or of the project may be identified according to various parameters including, density of the plots, ages, species of the trees, vegetation etc. The classification engine may apply a respective model to various plots of a project according to the most suitable model for the type of plot of the project. The classification engine 109 may apply data belonging to an area stretching across several plots having similar structures and characteristics to a model instead of applying each plot separately to a respective model to improve processing efficiency. As indicated above, a suitable model quantifies an amount of CO2 sequestered and / or estimates the CO2 absorption capacity of each plot for the remainder of the lifetime of the plot. Lidar data sets for the project being registered can be examined remotely by the classification engine 109 and / or uploaded on the software platform 10 for analysis to improve data processing efficiency. To obtain accurate data set relating to a given project, data can be collected from databases comprising field measurements as well as from available Lidar data sets. In an example the registration of a project on the software platform 10 and / or the processing by the classification engine 109 triggers gathering additional data via external databases 111 and / or via real-time data acquisition. Real time acquisition of data relating to a project can be achieved via Lidar, imaging techniques, Infrared, IR, sensing techniques etc. The external databases 111 may contain measured volumes of CO2 sequestered by a given project of part of it. The classification engine gathers this data to determine a CO2 absorption capacity of the project. Having gathered the relevant detailed data about the project, the classification engine determines a characteristic of the project based on the data captured. The characteristic of the project is indicative of a capacity of the project to avoid emission of or remove a greenhouse gas, such as CO2 from the atmosphere. In an example, the capture module obtains the CO2 absorption capacity of a project from the data entered via a user interface, which accurately represent the precise value of the CO2 absorption capacity of the project. In this case, the validation / verification module and / or the classification engine will verify the amount provided without modification or with minimal modification. The input injected in the classification engine 109 may also comprise a combination of outputs of the classification engine, and captured data and / or combination of derivatives of such data. Training a machine learning model of the Classification engine 109 involves applying different projects to specific models of CO2 absorption with specific parameters and comparing the CO2 volumes absorbed by the projects using the selected models with real field measurements. It further comprises assessing the differences or errors between the two and adjusting the parameters of the respective models to reduce the errors. Training the machine leaning results in a trained model whereby one or more suitable models having suitable parameters are selected for different projects according to their characteristics. The data entered through software platform over time can be used to train the machine learning model of the Classification engine 109. In an example, determining by the trained model of the classification engine 109 the characteristic of the project comprises determining the data associated with the project against data provided by a standard protocol. The data captured through the platform relative to a standard set of data will influence the determination of the characteristic of the project. Accordingly, classifying the project by the classification engine 109 is further based on the determined characteristic of the project relative to a standard. The trained model may be implemented using the application of an artificial neural network (ANN) or a convolutional neural network (CNN), which is trained as set out above. ANNs can be hardware - (neurons are represented by physical components) or software-based (computer models) and can use a variety of topologies and learning algorithms. ANNs usually have at least three layers that are interconnected. The first layer consists of input neurons. Those neurons send data on to the second layer, referred to a hidden layer which implements a function and which in turn sends the output neurons to the third layer. There may be a plurality of hidden layers in the ANN. With respect to the number of neurons in the input layer, this parameter is based on training data. The second or hidden layer in a neural network implements one or more functions. For example, the function or functions may each compute a linear transformation or a classification of the previous layer or compute logical functions. For instance, considering that the input vector can be represented as x, the hidden layer functions as h and the output as y, then the ANN may be understood as implementing a function fusing the second or hidden layer that maps from x to h and another function g that maps from h to y. So the hidden layer’s activation is f(x) and the output of the network is g(f(x)). CNNs can also be hardware or software based and can also use a variety of topologies and learning algorithms. A CNN usually comprises at least one convolutional layer where a feature map is generated by the application of a kernel matrix to an input image. This is followed by at least one pooling layer and a fully connected layer, which deploys a multilayer perceptron which comprises at least an input layer, at least one hidden layer and an output layer. The at least one hidden layer applies weights to the output of the pooling layer to determine an output prediction. Standards for the voluntary carbon market have been specified. The Emission Trading System, ETS is a regulated carbon market which defines targets to reduce net greenhouse gas emissions. It defines the volumes of greenhouses gases such as carbon, methane, fluorinated gases and ozone-depleting substances that need to be removed from the atmosphere. The VCM allows parties around the world to emit allowances (e.g carbon credits) that are purchased by emitters of greenhouse gases (e.g corporations) to reduce global greenhouse gas emissions. Several organizations such as the world bank and IOSCO have been developing a set of standards for the VCM. These standards propose a harmonization of a definition of a project around the world. The Integrity Council for the Voluntary Carbon Credit Market, ICVCM, is an independent governance body, which defined a framework to set a voluntary global threshold standard for quality in the voluntary carbon market. This framework is to help buyers recognize and put a price on high-integrity, high quality carbon credits. As indicated above, the results or output of the classification engine 109 is recorded to the blockchain 115. In an implementation, the results of the classification engine as well as the data captured relating to the carbon reduction project are recorded to the blockchain. In a further implementation, the output of the classification engine as well as the data captured, and the input data injected into the classification engine are recorded to the blockchain. The system comprising the software platform and its on-boarding process together with the blockchain infrastructure ensures that the project is real, legitimate and that its CO2 absorption capacity derived by the classification engine is accurate and reliable. It gives the public including any rating agency active on the VCM a degree of confidence regarding the legitimacy and quality of issued carbon credits associated with the projects. The described system facilitates projects compliance with standards which require that each carbon credit must legitimately measure a tonne of CO2 absorbed or avoided. The software platform 10 also allows registration of a project for a given period of time namely a registration period. Accordingly, a registration of a project may be renewed at, before or after the expiry of the registration period. A renewal of a registration of a project is intended to extend the registration of the project. The software platform 10 allows registration of operational projects such that when the project becomes exhausted or ceases to be operational, it can be deregistered from the platform 10. When a project is deregistered from the platform, the Token is automatically de-allocated from the project. In an example, when a project is deregistered from the platform, its Token Lifetime is forced to expire so that the Token associated with a deregistered project is no longer usable and / or tradeable. In an example, when a Token expires as its Lifetime expires, it is returned to a pool of pre-generated Tokens reserved for given geospatial locations until it is claimed for a project or allocated to a project. When a person starts a new project at a location where a previous project was terminated, the new project at that location may be registered with the Token which is allocated to that physical location. As indicated before, the Classification Engine obtains the exact amount of CO2 sequestered by a project since its creation or the amount of CO2 expected to be sequestered by the project for the remainder of its lifetime. In an example, the Token allocated to a project is associated with a Token Lifetime, so that the Token expires upon expiry of the Token Lifetime. The Token Lifetime is attributed to the Token before, during or after the registration of the project. The Token Lifetime may be comprised within the Token itself or be linked to the Token in a data structure. In a further example, the Token Lifetime attributed to the Token is given the value of the remainder of the lifetime of the project so that the Token expires before or at the end of the life of the project. This ensures that a Token identifying a project at a location is not valid anymore when the project has already ended its lifecycle. This prevents the Token of a project whose lifecycle has ended to be spent on the market, protecting carbon producers against buying a Token of a project that is no longer operational or in existence. It is noted that the smart contracts can verify the Token Lifetime before executing to prevent that a Token whose lifetime is expired is subject of a trade. It is noted that smart contracts execution can effect the trades of Tokens allocated to projects or of corresponding carbon credits. When the Token Lifetime indicates the remaining lifetime of a project, it affects the market price of the Token or corresponding carbon credit(s) since the value of a project is dependent on its remaining lifespan. When the remaining lifespan of a project is not known, it can be replaced by a configured registration period which is a period during which the registration of a project on the software platform 10 is valid. In some examples, the Token Lifetime carries the value of the registration period. The Token Lifetime decreases over time such that it expires substantially simultaneously with the expiry of the registration of the project on the platform 10. The Token Lifetime achieves the same purpose as above of affecting market price and ensuring that Tokens whose projects registration have expired are not traded on the market. Registration of a project on the software platform 10 gives the public the guarantee that the project will certainly be valid and in existence for at least the remainder of the registration period. When the remaining Lifetime of the project is known, a combination of the remaining Lifetime of the project and the registration period of the project can be encoded within the Token Lifetime. Figure 5 illustrates a stack showing different layers across which the system disclosed operates. It comprises an Enrichment Layer 510 which specifies the determination of a characteristic of a project, which enables a third party to appreciate the value of a project in terms of its CO2 absorption capacity. An Asset Layer 520 is defined, which comprises different types of projects including Biological Assets and Mechanical Assets. A Volumetric Unit 530 is a layer that comprises a set of definition captured from various standards. The Volumetric Unit can gather various information relating to the volumes of CO2. For example, it can gather and update the volume of CO2 that standard organisations and policy makers consider should be sequestered from the atmosphere by a target date. It is estimated that there is about 50 Gigatonne of carbon that needs to be removed from the atmosphere by 2050. The Volumetric Unit can also store various standard recommended volumes of CO2 absorption for different types of projects according to different parameters including sizes, location etc. The Volumetric Unit can be used to assess the capacity of each project to absorb CO2 as a percentage of a volume recorded in a standard. This information can be further used to determine the characteristics of the project impacting de facto a market price of a carbon credit of the project. A Digital Addressing layer 540 operates on different interfaces through which the software platform 10 communicates. For instance, IPv6 addresses can be used to exchange data between the software platform 10 and any trusted third party 107 and / or sources 111 and / or an addressable entity located on the project itself. A blockchain address such as a Unique Bitcoin address can be used by the software platform 10 to communicate with a selected blockchain 115. A base layer 550 presents the location of a project on which is based a Token as well as a global legal Entity Identifier for the project. The location of a project can be captured using service providers such as What3Words ® Address. The Global Legal Identifier of a project can be captured using services such as the Global Legal Entity Identifier Foundation (GLEIF). As indicated above, FIGURE 4 shows an example of classification comprising several types of projects including biological and mechanical projects. A project can include extracting methane gas and converting it into a solid and deposit it at a location such as an old un-used mine. Examples of projects include a transformation project using graphene to capture carbon from the air. Biological carbon capture projects comprise one or more biological elements capable of capturing and storing GHG to absorb or to sequester GHG from the air. For instance, a biological element may be a tree. A project may include one type of trees or several types of trees. Any given project would include one or more trees of a given type. A carbon capture project may comprise other biological elements including but not limited to a plant, an agricultural element, a living organism such as a microbe or a bacteria, a fungus, an insect such as a bug. Each biological element has an intrinsic capability to absorb and remove CO2 from the atmosphere. Trees and green plants in general have the capability to absorb more CO2 than they reject into the atmosphere through photosynthesis. The net absorption capability of a tree or a green plant depends on their specific features. The species, the age, the weight, the height, the diameter of the trunk of the tree etc may impact CO2 absorption efficiencies. Mangrove trees and other marshy tree systems that have their roots based in waterways tend to have high carbon sequestration abilities. Other biological projects can comprise natural elements such as Peatlands which have an important ability to sequester carbon from the atmosphere. Peatlands are seen as excellent carbon sinks and are candidates to be registered on the software platform as long-term carbon capture projects. The systems and methods of the described examples may be implemented in a computer program product such as a physical, non-transitory computer readable medium having computer readable instructions. The medium may be provided in various forms, including one or more diskettes, compact disks, tapes, chips, magnetic and electronic storage media, volatile memory, non-volatile memory and the like. Non-transitory computer-readable media may include all computer-readable media, with the exception being a transitory, propagating signal. The term non-transitory is not intended to exclude computer readable media such as primary memory, volatile memory, RAM and so on, where the data stored thereon may only be temporarily stored. The computer useable instructions may also be in various forms, including compiled and uncompiled code. Although numerous details are provided through the above examples, other examples may be practiced without these details. The description is not to be considered as limiting the scope of the disclosure.

Claims

1. A computer implemented method for classifying an environmental project, the method implemented by one or more processors associated with a platform for managing environmental projects, the method comprising the steps of:obtaining project data associated with a project, the project being an emission avoidance project or an emission reduction project;determining a characteristic of the project based on one or more attributes of the project data that relates to a feature of interest in a data set associated with a data model trained for classifying environmental projects;wherein the determined characteristic is indicative of a capacity of the project to avoid emission of or remove a greenhouse gas, GHG, from the air;assigning the determined characteristic to the project.

2. The method as claimed in claim 1 further comprising, training the data model to classify environmental projects including the steps of: providing a plurality of data sets, each pertaining to a type of environmental project;for a given data set,providing a training input including a label for a feature of interest associated with the given data set; andproviding a training output identifying a characteristic that is associated with the feature of interest pertaining to the label.

3. The method as claimed in 2 further including:providing input project data to the trained data model;the method further comprises, implementing using the trained data model the steps of:detecting at least one feature in the input project data that corresponds to a feature of interest in the plurality of data sets;determining that the at least one feature in the input project data is associated with a label in the training input that corresponds to said feature of interest;identifying that the input project data is of the characteristic associated with said label; andproviding as an output the type of environmental project based on the identified characteristic.

4. The method as claimed in any one of claims 1 to 3, wherein based on one or more attributes of the project data, selecting a trained data model amongst a plurality of data models trained for classifying environmental projects; anddetermining the characteristic of the project by providing the project data to the selected trained data model.

5. The method as claimed in claim 4, wherein obtaining project data comprises obtaining a first data; andbased on the selected trained model, obtaining a second data from a trusted source of information; anddetermining the characteristic of the project by providing the first data, and the second data to the selected trained data model.

6. The method as claimed in any one of claims 1 to 5, wherein obtaining project data comprises verifying the project data by means of a validation module; anddetermining the characteristic of the project by providing the verified data to the trained data model.

7. The method as claimed in any one of claims 1 to 6, wherein the project has an expected Lifecycle, and wherein the project data comprises a period of the expected Lifecycle of the location.

8. The method as claimed in any one of claims 1 to 7, wherein obtaining project data associated with the project comprises:deriving further data associated with the project to provide enhanced data; and determining the characteristic of the project based on the data including the enhanced data.

9. The method as claimed in any one of claims 1 to 8, wherein obtaining project data comprises deriving a location data of the project using a location identification technique such as GPS location tracking technique.

10. The method as claimed in claim 9, wherein the location data comprises a data set defining a geographic space surrounding the location.

11. The method as claimed in any one of claims 9 to 10, wherein the location data comprises at least a geospatial data set.

12. The method as claimed in any one of claims 9 to 11, wherein the location data is captured via a digital map which shows the physical location of the project.

13. The method as claimed in any one of claims 1 to 12, wherein the data associated with the project comprises a metadata associated with a geographical location pertaining to said project; the method including capturing the metadata associated with the project from a digital image of the location.

14. The method as claimed in any one of claims 1 to 13, wherein the project is associated with a biological element capable of capturing and storing GHG, so as to absorb or to sequester GHG from the air, the biological element comprising at least one of a plant, an agricultural element, a tree, a living organism such as a microbe or a bacteria, a fungus, an insect such as a bug.

15. The method as claimed in any one of claims 1 to 13, wherein the project is associated with a mechanical element capable of capturing and storing GHG, so as to absorb or to sequester GHG from the air, wherein the mechanical element comprises a material including at least one of graphene, concrete mixtures containing GHG absorbing ingredients.

16. The method as claimed in any one of claims 1 to 15, wherein the project is associated with at least one of a piece of land, an amount of a mineral, a liquid such as an aqueous solution capable of removing GHG from the air.

17. The method as claimed in any one of claims 1 to 16, wherein determining the characteristic of the project comprises determining a volume or a mass of GHG expected to be captured by a location associated with the project during a period of an expected Lifecycle of the project.

18. The method as claimed in any one of claims 1 to 17, wherein the GHG is carbon dioxide, the method further comprising using the determined characteristic of the project for issuing or updating a rating to a carbon credit associated with the project; or issuing or updating a carbon credit associated with the project.

19. The method as claimed in any one of claims 1 to 18 further comprising, storing the determined characteristic as metadata associated with the project, said metadata being stored on a blockchain or on an off-block storage resource.

20. The method as claimed in any one claims 1 to 19, wherein the project is represented by a digital token for a given geographical location, and the digital token is associated with a Lifetime of the project; wherein the token expires upon expiry of the Lifetime.

21. The method as claimed in claim 20, wherein the token is associated with an output of a blockchain transaction.

22. A method as claimed in any one of claims 20 to 21, further comprising: determining whether the project is associated with an existing token;wherein if the project is associated with an existing token;linking the existing token to the determined characteristic and wherein if the project is not associated with an existing token;generating a token for the project based on the project data.

23. The method as claimed in claim 22, wherein generating the token comprises extracting a geographical location and an identity of an entity associated with the project from the project data;allocating the extracted location, the identity and determined characteristic associated with the project to the token.

24. A method of registering a project, wherein the project is associated with a plurality of locations, the method comprising:capturing data associated with the plurality of locations;allocating a project token to each location among the plurality and to the identity of an entity associated with project and / or given location;providing the captured data into a trained model;determining, using the trained model, a characteristic of the project based on the captured data associated with the plurality of location;wherein the characteristic of the project is indicative of a capacity of the project to avoid or remove a greenhouse gas, GHG, from the atmosphere;assigning the characteristic of the project to the allocated project token.

Images