Hurricane Rainfall Prediction

Modeling and Simulation & Scientific Computing

ABOUT

This project uses two different models to predict the total amount of rainfall over time along the path of Hurrican Fran in 1996. The models used are based on "Evaluation of GFDL and Simple Statistical Model Rainfall Forecasts for U.S. Landfalling Tropical Storms" by Tuleya et al. Specifically, the project implements the rainfall climatalogy and persistence (R-CLIPER) and the Tropical Rainfall Measuring Mission (TRMM) models defined in the aforementioned paper, applied to the path of Hurricane Fran in 1996. The goal for this project is to accurately replicate the results of the aforementioned models in comparison with Tuleya et al.'s results with clear and logical graphics.

  • TECHNOLOGIES
  • Python
  • matplotlib
  • sympy
  • numpy
  • cartopy

SKILLS USED

  • Design models based on previous work
  • Model and simulate rainfall over time and distance
  • Validate model performance against previous models and data
  • Verify behavior of the models based on published paper
TRMM model prediction
R-CLIPER model prediction

Comparing the results of both models to the benchmark, the gauge-based R-CLIPER and the TRMM model both produce results with the same trend of rainfall, while also being within the same magnitude of rainfall amount. While the rainfall range in inches more closely reflects the amount of rainfall in the benchmark, both models produce results that match those of the becnhmark. Consequently, both of the models could be used in the future in order to detect the amount of rainfall along a given storm's path, helping to prepare those in the impacted regions before they face the brunt of a storm.