Augmenting climate impact and hazard models

1 . A computer-implemented method for augmenting a plurality of climate impact and hazard models, the method comprising:

modifying, by one or more processors, the plurality of climate impact and hazard models based on a user-specified requirement and at least one intelligent workflow of intelligent workflows, wherein modifying the plurality of climate impact and hazard models comprises at least one selected from a group consisting of: onboarding, managing, and scaling each of the plurality of climate impact and hazard models based on the user-specified requirement and the at least one intelligent workflow and

wherein a framework for the intelligent workflows comprises a plurality of framework components and wherein the plurality of framework components includes an API (application programming interface), a workflow specification layer, data and model ontology, workflow orchestration layer, model validation layer, data storage and interchange layer, workflow run and data sources and

wherein the intelligent workflows are a sequence of processes for augmentation of the climate impact and hazard models and the intelligent workflows can be configured to adapt to the climate impact and hazard models and type of modification that is to be performed;

identifying, by the one or more processors, for each of the modified plurality of climate impact and hazard models, at least one model parameter associated with a respective modified climate impact and hazard model, by processing the respective modified climate impact and hazard model with a model enhancing machine learning model;

determining, by the one or more processors, for each of the modified plurality of climate impact and hazard models, at least one operation based on the user-specified requirement, the at least one intelligent workflow, and the at least one model parameter corresponding to the respective modified climate impact and hazard model; and

executing, by the one or more processors, for each of the modified plurality of climate impact and hazard models, the at least one operation corresponding to the respective modified climate impact and hazard model, based on the at least one intelligent workflow.

2 . The computer-implemented method of claim 1 , further comprising unifying the modified plurality of climate impact and hazard models.

3 . The method of claim 1 , wherein modifying the plurality of climate impact and hazard models includes onboarding each of the plurality of climate impact and hazard models by generating and providing geospatial data interfaces, pre-processing, post-processing, and validation tools.

4 . The computer-implemented method of claim 1 , further comprising storing the modified plurality of climate impact and hazard models on a hybrid cloud environment.

5 . The computer-implemented method of claim 4 , further comprising:

storing, by the one or more processors, the modified plurality of climate impact and hazard models, wherein storing the modified plurality of climate impact and hazard models includes managing, consolidating, versioning, and benchmarking each of the modified plurality of climate impact and hazard models.

6 . The computer-implemented method of claim 1 , wherein identifying the at least one model parameter associated with the respective modified climate impact and hazard model is further based on ontologies and knowledge graphs corresponding to the respective modified climate impact and hazard model.

7 . The computer-implemented method of claim 1 , wherein the at least one operation includes at least one selected from the group consisting of: model calibration, uncertainty quantification, model validation, localization and scaling of the modified plurality of climate impact and hazard model.

8 . The computer-implemented method of claim 1 , wherein the at least one operation includes model calibration, and executing the operation comprises:

generating, by the one or more processors, a ground truth test data;

inputting, by the one or more processors, the ground truth test data to a ground truth machine learning model;

generating, by the one or more processors, a predicted output, based on inputting the ground truth test data to the modified plurality of climate impact and hazard model; and

adapting, by the one or more processors, parameters of the modified plurality of climate impact and hazard model based on a comparison between the ground truth machine learning model and predicted output.

9 . The computer-implemented method of claim 1 , wherein the at least one operation includes uncertainty quantification, and executing the at least one operation comprises learning a joint distribution of data input to the modified climate impact and hazard model, and uncertainty of the modified climate impact and hazard model.

10 . The computer-implemented method of claim 1 , wherein the at least one operation includes model validation, and executing the at least one operation comprises: validating the modified plurality of climate impact and hazard model based on the user-specified requirement.

11 . The computer-implemented method of claim 1 , wherein the at least one operation includes scaling, and executing the at least one operation includes adapting the at least one model parameter of the modified plurality of climate impact and hazard model based on the user-specified requirement.

12 . The computer-implemented method of claim 1 , wherein the modified plurality of climate impact and hazard models includes at least one selected from a group consisting of: flooding models, wildfire models, drought models, rainfall models, heat wave models, and cold wave models.

13 . The computer-implemented method of claim 1 , wherein the user-specified requirement includes at least one selected from a group consisting of: a geographical location, a time period, and a modelled hazard.

14 . A computer system for augmenting a plurality of climate impact and hazard models, the system comprising:

one or more computer processors;

at least one computer-readable storage medium, and program instructions stored on the at least on computer readable storage medium, the computer system further comprising:

a modelling unit configured with program instructions to modify the plurality of climate impact and hazard models based on a user-specified requirement and at least one intelligent workflow of intelligent workflows, wherein modifying the plurality of climate impact and hazard models comprises at least one selected from a group consisting of: onboarding, managing, and scaling each of the plurality of climate impact and hazard models based on the user-specified requirement and the at least one intelligent workflow and

wherein a framework for the intelligent workflows comprises a plurality of framework components and wherein the plurality of framework components includes an API (application programming interface), a workflow specification layer, data and model ontology, workflow orchestration layer, model validation layer, data storage and interchange layer, workflow run and data sources and

wherein the intelligent workflows is a sequence of processes for augmentation of the plurality climate impact and hazard models and the intelligent workflows can be configured to adapt to the plurality climate impact and hazard models and type of modification that is to be performed;

a model analysis unit configured with program instructions to identify, for each of the modified plurality of climate impact and hazard models, at least one model parameter associated with a respective modified climate impact and hazard model, by processing the respective modified climate impact and hazard model with a model enhancing machine learning model;

an operation identification unit configured with program instructions to determine, for each of the modified plurality of climate impact and hazard models, at least one operation based on the user-specified requirement, the at least one intelligent workflow, and the at least one model parameter corresponding to the respective modified climate impact and hazard model; and

a model adaptation unit configured with program instructions to execute, for each of the modified plurality of climate impact and hazard models, the at least one operation corresponding to the respective modified climate impact and hazard model, based on the at least one intelligent workflow.

15 . The computer system of claim 14 , wherein modifying the plurality of climate impact and hazard models includes onboarding each of the modified plurality of climate impact and hazard models by generating geospatial data interfaces, and providing pre-processing, post-processing, and validation tools.

16 . The computer system of claim 14 , further comprising:

storing, by the one or more processors, the modified plurality of climate impact and hazard models, wherein storing the modified plurality of climate impact and hazard models includes managing, consolidating, versioning, and benchmarking each of the modified plurality of climate impact and hazard models.

17 . The computer system of claim 14 , wherein identifying the at least one model parameter associated with the respective modified climate impact and hazard model is further based on ontologies and knowledge graphs corresponding to the respective modified climate impact and hazard model.

18 . The computer system of claim 14 , wherein the at least one operation includes at least one selected from the group consisting of: model calibration, uncertainty quantification, model validation, localization and scaling of the modified plurality of climate impact and hazard model.

19 . The computer system of claim 14 , wherein the at least one operation includes model calibration, and executing the operation comprises:

program instructions to generate a ground truth test data;

program instructions to input the ground truth test data to a ground truth machine learning model;

program instructions to generate a predicted output, based on inputting the ground truth test data to the modified plurality of climate impact and hazard model; and

program instructions to adapt the at least one parameter of the respective modified climate impact and hazard model, based on a comparison between the ground truth machine learning model and predicted output.