[GSoC Project Proposal]: Land Cover Integration for Parametric Hurricane Model (PaHM)

[AtiehAlipour-NOAA]

Project Description

This project aims to enhance the Generalized Asymmetric Holland Model (GAHM) within the Parametric Hurricane Modeling System (PaHM) framework by incorporating land cover effects on wind speeds during tropical cyclones. GAHM is a parametric model that provides realistic representations of wind and pressure fields using track file data from various sources (National Hurricane Center, Hurricane Databases, International Best Track Archive for Climate Stewardship). While GAHM effectively models wind fields, the current standalone versions don't account for varying land cover characteristics that significantly affect wind speeds over land.
The project will focus on:

• Implementing efficient access and storage of land cover datasets (NLCD, NOAA C-CAP)
• Developing algorithms to calculate roughness values based on upwind land cover characteristics
• Modifying wind fields using boundary layer theory to account for land cover effects
• Integration of these features into the PaHM framework

This enhancement will improve the accuracy of wind field predictions over land, which is crucial for storm surge and coastal flooding forecasts. The implementation will build upon existing successful approaches used in the Advanced Circulation (ADCIRC) model.

Software:
Parametric Hurricane Modeling System (PaHM): https://github.com/noaa-ocs-modeling/PaHM

Expected Outcomes

• A new version of GAHM in PaHM that incorporates land cover effects on wind fields, including:
• Efficient handling of land cover datasets
• Implementation of roughness calculations
• Modified wind field calculations accounting for land cover types
• Documentation and validation of the new features

Skills Required

• Fortran (required for PaHM/GAHM development)
• Python for analysis and visualization; Libraries: Xarray, Cartopy, Matplotlib, NetCDF4 for handling and plotting gridded data
• Some experience with numerical modeling.

Additional Background/Issues

No response

Mentor(s)

Rick Luettich (rick_luettich@unc.edu ), Mansur Jisan (mansur.jisan@noaa.gov )

Expected Project Size

350 hours

Project Difficulty

Intermediate

[Srajald]

@AtiehAlipour-NOAA
Hi,
I am Srajal, a B. Tech undergraduate final year student at NIT Jalandhar, India.
I have gone through this project and was doing a local setup of your github repo..
I am very much interested in your project, as this aligns with my core interest using ML/AI implementations..
I would like to contribute to this project as a GsoC contributor, 2025..
If there are any suggestions , then I would like to take them as a lesson...
Thanks

[srodif]

Hello!
I would also be interested in this project, what is the preferred contact method for the mentors?

There is one main thing I would love a clarification on, I noticed that the project will also have a scientific focus in parts (calculating roughness, wind fields, Fortran usage). I have recently contributed to a CASSINI Hackathon for satellite imagery usage and it was apparent how helpful and important domain specific knowledge was. Is there a way to support people like me, with a main focus in software engineering (but quite an interest in scientific applications!), in finding and learning (important papers, methods) about the specific domain?

[mansurjisan]

@AtiehAlipour-NOAA Hi, I am Srajal, a B. Tech undergraduate final year student at NIT Jalandhar, India. I have gone through this project and was doing a local setup of your github repo.. I am very much interested in your project, as this aligns with my core interest using ML/AI implementations.. I would like to contribute to this project as a GsoC contributor, 2025.. If there are any suggestions, then I would like to take them as a lesson... Thanks

Hi Srajal,

Thanks for your interest in the "Land Cover Integration for PaHM" project — we're glad to hear you're exploring the repository and thinking about contributing!

Just to clarify, this project focuses on enhancing a physics-based parametric hurricane wind model by integrating land cover data to better represent surface roughness effects. It does not involve machine learning or AI techniques, but rather requires a good grasp of numerical modeling, boundary layer theory, and working with Fortran and Python.

That said, if you're still excited about working on land surface impacts on hurricane wind fields and are comfortable learning or working with Fortran and geospatial datasets like NLCD, we’d love to see your continued engagement.

We recommend the following next steps:

• Familiarize yourself with the PaHM codebase
• Review documentation on surface roughness and land cover impacts on wind (you can start with literature from the ADCIRC model)
• Try to reproduce a simple PaHM run and visualize the wind field using Python

Please feel free to contact us via email if you need additional information!

Best,
Mansur

[mansurjisan]

Hello! I would also be interested in this project, what is the preferred contact method for the mentors?

There is one main thing I would love a clarification on, I noticed that the project will also have a scientific focus in parts (calculating roughness, wind fields, Fortran usage). I have recently contributed to a CASSINI Hackathon for satellite imagery usage and it was apparent how helpful and important domain specific knowledge was. Is there a way to support people like me, with a main focus in software engineering (but quite an interest in scientific applications!), in finding and learning (important papers, methods) about the specific domain?

Hi @srodif ,

Thanks for your interest, and it's great to hear about your experience with the CASSINI Hackathon. You're absolutely right that domain-specific knowledge plays a significant role in scientific modeling projects like this one, and I really appreciate that you're thinking ahead about it.

You're not alone — many successful GSoC contributors come from software engineering backgrounds and gradually build up the scientific side, especially with mentorship and the right resources. For this project, the core challenges lie at the intersection of surface boundary layer physics, wind modeling, and numerical implementation. Your strong software foundation is a great starting point.

To help you begin exploring the domain, here are a few suggested resources:

• Introduction to Boundary Layer Meteorology by Roland Stull — especially the sections on surface roughness and drag

• Powell et al. (2003) – “Reduced drag coefficient for high wind speeds in tropical cyclones” — a useful paper for understanding wind roughness concepts

• PaHM Wind Model Documentation — this is the modeling framework we’ll be working with to integrate land cover effects

• NLCD 2019 Land Cover Data — to get familiar with the type of surface data used in the project

Please feel free to go over these materials. I’d be happy to discuss any of the concepts further or point you to more specific readings as needed.

You're welcome to continue the discussion here in the GSoC issue thread, but if you need additional information or have more detailed questions, don’t hesitate to reach out to me via email at mansur.jisan@noaa.gov.

If you're considering submitting a proposal, you can try a basic run of the model and engage with the repository through small contributions or questions.

Looking forward to hearing more from you.

Best,
Mansur