Binata Roy

University of Virginia

 Recent Activity

ABSTRACT:

This resource includes the input data for LSTM and seq2seq LSTM surrogate models for multi-step-ahead street-scale flood forecasting in Norfolk, VA, USA. The data consists of topographic features: topographic wetness index (TWI), depth to water (DTW) and elevation, and environmental features: hourly rainfall and tide level from gauge stations and water depth generated by the physics-based model TUFLOW.

There are three folders in this resource -
1. The "OriginalData" folder includes the CSV files for the top 20 daily storm events from 2016-2018 for the streets of Norfolk.
2. The "FloodproneStreets" folder includes shapefiles of the street segments (polygons of 7.2 m width x 50 m length) of Norfolk. Alongside, it includes a CSV file containing 22 flood-prone streets selected from the STORM report.
3. The "RelationalDatabase" folder includes three CSV files for node_data (varied spatially), tide_data (varied temporally) and weather_data tide_data (varied spatially and temporally) for efficient data management. The notebook script "create_relational_data.ipynb" is used to convert "OriginalData" to "RelationalDatabase".

The Python script of the LSTM and seq2seq LSTM surrogate models is available on GitHub https://github.com/br3xk/LSTM-and-seq2seq-LSTM-surrogate-models-for-street-scale-flood-forecasting
The output of the model is forecasted hourly water depth on the 22 flood-prone streets with 4-hr and 8-hr lead.

Show More

ABSTRACT:

This HydroShare resource was created to show performance test using the different reproducible approaches in this paper.

Show More

ABSTRACT:

Sciunit (https://sciunit.run/) is a tool that encapsulates a set of executions into an isolated, independent container. It allows computational scientists to create research objects, which can be reused and transferred to other computational environments for reproducibility. Sciunit containerizes a program by capturing the trace of its execution using system utilities. It stores the sequence of instructions to run the program and the input and output data content used by that program. Programs in this self-contained sandbox are reproduced on the system or transported to another system for re-execution.

In this resource, users can show how to reproduce a Sciunit Container that encapsulates MODFLOW-NWT Use Case in the James River watershed upstream of Richmond, VA, USA

Show More

ABSTRACT:

This is a collection resource for "Comparing containerization-based approaches for reproducible computational modeling of environmental systems" manuscript in Environmental Modeling and Software Journal.
HS-1: Collection Resource for Comparing Approaches to Achieve Reproducible Computational Modeling for Hydrological and Environmental Systems

For SUMMA simulation, we created two SUMMA model instances.
HS-2. Model Instance for the Impact of Stomatal Resistance Parameterizations on ET of SUMMA Model in Aspen stand at Reynolds Mountain East
HS-3. Model Instance for the Impact of Lateral Flow Parameterizations on Runoff of SUMMA Model at Reynolds Mountain East

Then there is a HS resource for reproducible approaches in local computational environments.
HS-4. A Virtual Box image that includes five local approaches:
- Approach-1 Compiling the core model software
- Approach-2 Containerizing the core model software only with Docker
- Approach-3 Containerizing all software with Docker
- Approach-4 Containerizing all software with Singularity
- Approach-5 Containerizing all software and modeling workflows with Sciunit

Next, there are four HS resources and a GitHub repository for reproducible approaches in remote computational environments.
HS-5. Approach-6 Using CUAHSI JupyterHub
HS-6. Approach-7 Using CyberGIS-Jupyter for water
HS-7. Approach-8 Using Sciunit in CUAHSI JupyterHub
HS-8. Approach-9 Using Sciunit in CyberGIS-Jupyter for water
Git-1. Approach-10 Using Binder (https://github.com/uva-hydroinformatics/SUMMA_Binder.git)

Lastly, we created a notebook for performance tests using the different reproducible approaches.
HS-9. Jupyter notebook for performance test using the different reproducible approaches

For additional description, we created two GitHub repositories to show how to create Docker and Singularity image for Approach-2,3, and 4.
Git-2. Description of Approach-3 to show how to create Docker environments (https://github.com/uva-hydroinformatics/SUMMA_Docker_Training.git)
Git-3. Description of Approach-4 to show how to use a Singularity image (https://github.com/uva-hydroinformatics/SUMMA_Singularity_In_Rivanna.git)

As a result, we shared a created Singularity image for a model program resource.
HS-10: A singularity image for the reproducibility of SUMMA modeling

Show More

ABSTRACT:

This HydroShare resource provides five local approaches using Virtual Box image.
First, users need to install Virtual Box (https://www.virtualbox.org/wiki/Downloads) at first.
Then import this "research.ova" to create a SUMMA and pySUMMA computational environment in Virtual Box.

After creating "research.ova" image on Virtual Box, users need to move to the "/home/hydro/project/Performance_Test" folder to start SUMMA run.
Then, users can follow the "instruction.txt" in each approach foler.
The password of this "research.ova" image is "hydro."

This Virtual Box image five local approaches:
- Approach-1 Compiling the core model software
- Approach-2 Containerizing the core model software only with Docker
- Approach-3 Containerizing all software with Docker
- Approach-4 Containerizing all software with Singularity
- Approach-5 Containerizing all software and modeling workflows with Sciunit

Show More

 Contact

Mobile 4343654359
Email (Log in to send email)
Resources
All 0
Collection 0
Resource 0
App Connector 0
Resource Resource

ABSTRACT:

This HydroShare resource provides a Singularity image for Local Approach 4: Containerizing all software with Singularity for "Comparing containerization-based approaches for reproducible computational modeling of environmental systems" manuscript in Environmental Modeling and Software Journal.

For more detailed information, please see this GitHub
Git-3. Description of Approach-4 to show how to use a Singularity image (https://github.com/uva-hydroinformatics/SUMMA_Singularity_In_Rivanna.git)

Show More
Resource Resource

ABSTRACT:

This HydroShare resource provides the Jupyter Notebooks for the reproducibility of SUMMA modeling using CyberGIS-Jupyter for water in the manuscript of "Comparing Approaches to Achieve Reproducible Computational Modeling for Hydrological and Environmental Systems" in Environmental Modeling and Software.

To find out the instructions on how to run Jupyter Notebooks, please refer to the README file which is provided in this resource.

Show More
Resource Resource

ABSTRACT:

This HydroShare resource provides the Jupyter Notebooks for the reproducibility of SUMMA modeling using CUAHSI JupyterHub in the manuscript of "Comparing Approaches to Achieve Reproducible Computational Modeling for Hydrological and Environmental Systems" in Environmental Modeling and Software.

To find out the instructions on how to run Jupyter Notebooks, please refer to the README file which is provided in this resource.

Show More
Resource Resource

ABSTRACT:

This HydroShare resource provides the Jupyter Notebooks for the reproducibility of SUMMA modeling using Sciunit in CyberGIS-Jupyter for water in the manuscript of "Comparing Approaches to Achieve Reproducible Computational Modeling for Hydrological and Environmental Systems" in Environmental Modeling and Software.

To find out the instructions on how to run Jupyter Notebooks, please refer to the README file which is provided in this resource.

Show More
Resource Resource

ABSTRACT:

This HydroShare resource provides the Jupyter Notebooks for the reproducibility of SUMMA modeling using Sciunit in CUAHSI JupyterHub in the manuscript of "Comparing Approaches to Achieve Reproducible Computational Modeling for Hydrological and Environmental Systems" in Environmental Modeling and Software.

To find out the instructions on how to run Jupyter Notebooks, please refer to the README file which is provided in this resource.

Show More
Resource Resource

ABSTRACT:

This HydroShare resource provides five local approaches using Virtual Box image.
First, users need to install Virtual Box (https://www.virtualbox.org/wiki/Downloads) at first.
Then import this "research.ova" to create a SUMMA and pySUMMA computational environment in Virtual Box.

After creating "research.ova" image on Virtual Box, users need to move to the "/home/hydro/project/Performance_Test" folder to start SUMMA run.
Then, users can follow the "instruction.txt" in each approach foler.
The password of this "research.ova" image is "hydro."

This Virtual Box image five local approaches:
- Approach-1 Compiling the core model software
- Approach-2 Containerizing the core model software only with Docker
- Approach-3 Containerizing all software with Docker
- Approach-4 Containerizing all software with Singularity
- Approach-5 Containerizing all software and modeling workflows with Sciunit

Show More
Collection Collection

ABSTRACT:

This is a collection resource for "Comparing containerization-based approaches for reproducible computational modeling of environmental systems" manuscript in Environmental Modeling and Software Journal.
HS-1: Collection Resource for Comparing Approaches to Achieve Reproducible Computational Modeling for Hydrological and Environmental Systems

For SUMMA simulation, we created two SUMMA model instances.
HS-2. Model Instance for the Impact of Stomatal Resistance Parameterizations on ET of SUMMA Model in Aspen stand at Reynolds Mountain East
HS-3. Model Instance for the Impact of Lateral Flow Parameterizations on Runoff of SUMMA Model at Reynolds Mountain East

Then there is a HS resource for reproducible approaches in local computational environments.
HS-4. A Virtual Box image that includes five local approaches:
- Approach-1 Compiling the core model software
- Approach-2 Containerizing the core model software only with Docker
- Approach-3 Containerizing all software with Docker
- Approach-4 Containerizing all software with Singularity
- Approach-5 Containerizing all software and modeling workflows with Sciunit

Next, there are four HS resources and a GitHub repository for reproducible approaches in remote computational environments.
HS-5. Approach-6 Using CUAHSI JupyterHub
HS-6. Approach-7 Using CyberGIS-Jupyter for water
HS-7. Approach-8 Using Sciunit in CUAHSI JupyterHub
HS-8. Approach-9 Using Sciunit in CyberGIS-Jupyter for water
Git-1. Approach-10 Using Binder (https://github.com/uva-hydroinformatics/SUMMA_Binder.git)

Lastly, we created a notebook for performance tests using the different reproducible approaches.
HS-9. Jupyter notebook for performance test using the different reproducible approaches

For additional description, we created two GitHub repositories to show how to create Docker and Singularity image for Approach-2,3, and 4.
Git-2. Description of Approach-3 to show how to create Docker environments (https://github.com/uva-hydroinformatics/SUMMA_Docker_Training.git)
Git-3. Description of Approach-4 to show how to use a Singularity image (https://github.com/uva-hydroinformatics/SUMMA_Singularity_In_Rivanna.git)

As a result, we shared a created Singularity image for a model program resource.
HS-10: A singularity image for the reproducibility of SUMMA modeling

Show More
Resource Resource

ABSTRACT:

Sciunit (https://sciunit.run/) is a tool that encapsulates a set of executions into an isolated, independent container. It allows computational scientists to create research objects, which can be reused and transferred to other computational environments for reproducibility. Sciunit containerizes a program by capturing the trace of its execution using system utilities. It stores the sequence of instructions to run the program and the input and output data content used by that program. Programs in this self-contained sandbox are reproduced on the system or transported to another system for re-execution.

In this resource, users can show how to reproduce a Sciunit Container that encapsulates MODFLOW-NWT Use Case in the James River watershed upstream of Richmond, VA, USA

Show More
Resource Resource

ABSTRACT:

This HydroShare resource was created to show performance test using the different reproducible approaches in this paper.

Show More
Resource Resource

ABSTRACT:

This resource includes the input data for LSTM and seq2seq LSTM surrogate models for multi-step-ahead street-scale flood forecasting in Norfolk, VA, USA. The data consists of topographic features: topographic wetness index (TWI), depth to water (DTW) and elevation, and environmental features: hourly rainfall and tide level from gauge stations and water depth generated by the physics-based model TUFLOW.

There are three folders in this resource -
1. The "OriginalData" folder includes the CSV files for the top 20 daily storm events from 2016-2018 for the streets of Norfolk.
2. The "FloodproneStreets" folder includes shapefiles of the street segments (polygons of 7.2 m width x 50 m length) of Norfolk. Alongside, it includes a CSV file containing 22 flood-prone streets selected from the STORM report.
3. The "RelationalDatabase" folder includes three CSV files for node_data (varied spatially), tide_data (varied temporally) and weather_data tide_data (varied spatially and temporally) for efficient data management. The notebook script "create_relational_data.ipynb" is used to convert "OriginalData" to "RelationalDatabase".

The Python script of the LSTM and seq2seq LSTM surrogate models is available on GitHub https://github.com/br3xk/LSTM-and-seq2seq-LSTM-surrogate-models-for-street-scale-flood-forecasting
The output of the model is forecasted hourly water depth on the 22 flood-prone streets with 4-hr and 8-hr lead.

Show More