Pandey et al. (2025) Tropical cyclone heat potential monitoring and forecasting over the Indian ocean using the UM-based coupled model

Identification

• Journal: Theoretical and Applied Climatology
• Year: 2025
• Date: 2025-09-09
• Authors: Lokesh Kumar Pandey, Ankur Gupta, Imranali M. Momin, Akhilesh Mishra
• DOI: 10.1007/s00704-025-05762-y

Research Groups

• National Centre for Medium Range Weather Forecasting (NCMRWF), Noida, India
• Collaborations for NEMOVar development: Centre Européen de Recherche et Formation Avancée en Calcul Scientifique (CERFACS), European Centre for Medium-Range Weather Forecasts (ECMWF), Institut National de Recherche en Informatique et Automatique (INRIA), and the Met Office.

Short Summary

This study evaluates the performance of a Unified Model (UM)-based coupled atmosphere-ocean model in monitoring and forecasting Tropical Cyclone Heat Potential (TCHP) and its relationship with Sea Surface Height (SSH) over the Indian Ocean, demonstrating its utility for tropical cyclone intensification prediction. The model accurately simulates TCHP spatial patterns and maintains the physical link between TCHP and SSH, particularly in early forecasts, despite increasing biases at longer lead times.

Objective

• To evaluate the forecast performance of a UM-based coupled model in simulating Tropical Cyclone Heat Potential (TCHP) and its relationship with Sea Surface Height (SSH) during Cyclone Biparjoy (Arabian Sea, June 2023) and Cyclone Mocha (Bay of Bengal, May 2023).
• To assess the biases, correlation, and Root Mean Square Error (RMSE) of TCHP and SSH forecasts against analysis data.
• To provide insights into the model's operational forecast potential over the North Indian Ocean and its capacity to represent the oceanic thermal structure along cyclone tracks.

Study Configuration

• Spatial Scale: Global ocean model, focusing on the Indian Ocean (Arabian Sea and Bay of Bengal). Model resolution is 0.25° horizontally with 75 vertical levels. Specific evaluation regions: Arabian Sea (60°E–75°E, 5°N–25°N) and Bay of Bengal (85°E–100°E, 5°N–23°N).
• Temporal Scale: Daily ocean data from 2017 to 2023 for climatology and anomaly calculations. Forecasts extend up to 15 days. Case studies cover Cyclone Biparjoy (06–19 June 2023) and Cyclone Mocha (09–14 May 2023).

Methodology and Data

• Models used:
  • Atmospheric and Land Component: Unified Model (UM) and Joint UK Land Environment Simulator (JULES).
  • Ocean Component: Nucleus for European Modelling of the Ocean (NEMO).
  • Sea Ice Component: Los Alamos Sea Ice Model (CICE).
  • Coupler: OASIS coupler version 3.0.
  • Ocean Data Assimilation: NEMOVar (a variational data assimilation system based on NEMO), configured at ORCA025.
• Data sources:
  • Global daily ocean data from the NEMO model (2017–2023).
  • Temperature and salinity from the in-house NEMOVar assimilation system.
  • Daily TCHP and SSH forecasts from the NCMRWF operational coupled ocean-atmosphere prediction system.
  • NEMOVar analysis datasets (0.25° resolution) used as reference for forecast verification.
  • Best track data from the India Meteorological Department (IMD) for cyclone positions.

Main Results

• The Tropical Cyclone Heat Potential (TCHP) anomaly over the Arabian Sea (AS) was higher than the Bay of Bengal (BoB) during May and June 2023.
• The coupled model successfully simulates the spatial structure and magnitude of TCHP, particularly during early forecast days (Day-1 to Day-10), with low biases.
• Forecast biases for TCHP increase with lead time, showing overestimation after Day-10 over the AS and after Day-6 over the BoB.
• The model captures higher TCHP values along cyclone tracks, which are associated with a stronger potential for intensification.
• A strong positive correlation was observed between TCHP and Sea Surface Height (SSH): 0.71 during Cyclone Biparjoy (AS) and 0.78 during Cyclone Mocha (BoB).
• The coupled model successfully maintains this physical relationship between SSH and TCHP in both forecast and analysis fields, indicating its skill in capturing underlying ocean dynamics.
• Errors in the simulation of SSH were found to explain a large part of the biases in TCHP.
• Forecast Skill Metrics:
  • Arabian Sea (Biparjoy): TCHP bias remained relatively low up to Day-10, then gradually increased. SSH bias consistently increased with lead time. TCHP spatial correlation exceeded 0.86 throughout the 15-day forecast. SSH correlation decreased to 0.65 by Day-15. RMSE increased steadily for both.
  • Bay of Bengal (Mocha): TCHP bias increased significantly from Day-6 onward. SSH showed very little bias. TCHP spatial correlation remained above 0.95 for all forecast days. SSH correlation declined to 0.50 by Day-15. RMSE increased for both.

Contributions

• Demonstrates the robust performance of a UM-based coupled model for real-time Tropical Cyclone Heat Potential (TCHP) monitoring and forecasting over the Indian Ocean.
• Provides a comprehensive evaluation of the coupled model's forecast skill for TCHP and Sea Surface Height (SSH) up to 15 days, using bias, correlation, and RMSE metrics.
• Highlights the model's ability to accurately represent the physical relationship between SSH and TCHP, which is crucial for understanding upper-ocean heat content dynamics relevant to cyclone intensification.
• Offers valuable insights into the operational forecast potential of coupled modeling systems for improving tropical cyclone prediction and intensity monitoring in the North Indian Ocean.
• Emphasizes the critical role of accurate subsurface ocean heat content simulation by coupled models for enhancing tropical cyclone forecasts.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Citation

@article{Pandey2025Tropical,
  author = {Pandey, Lokesh Kumar and Gupta, Ankur and Momin, Imranali M. and Mishra, Akhilesh},
  title = {Tropical cyclone heat potential monitoring and forecasting over the Indian ocean using the UM-based coupled model},
  journal = {Theoretical and Applied Climatology},
  year = {2025},
  doi = {10.1007/s00704-025-05762-y},
  url = {https://doi.org/10.1007/s00704-025-05762-y}
}