Raghuvanshi et al. (2025) An Hourly Dataset of Moisture Budget Components Over the Indian Subcontinent (1940–2024)

Identification

• Journal: Scientific Data
• Year: 2025
• Date: 2025-11-11
• Authors: Akash Singh Raghuvanshi, Ankit Agarwal
• DOI: 10.1038/s41597-025-06044-y

Research Groups

• Department of Hydrology, Indian Institute of Technology, Roorkee, India

Short Summary

This paper introduces ERA5moistIN, a high-resolution (0.25° spatial, hourly temporal) dataset of atmospheric moisture budget components over the Indian subcontinent from 1940 to 2024, derived from ERA5 reanalysis. The dataset's physical consistency and reliability are validated against native ERA5 diagnostics, demonstrating its capability to accurately capture moisture transport dynamics, even during extreme events.

Objective

• To develop and validate a high-resolution, hourly dataset of atmospheric moisture budget components (ERA5moistIN) over the Indian subcontinent and surrounding ocean regions from 1940 to 2024, using ERA5 reanalysis data.
• To provide a physically consistent framework for diagnosing water cycle dynamics and assessing moisture transport processes, suitable for applications such as monsoon studies, extreme event attribution, and model evaluation.

Study Configuration

• Spatial Scale: Indian subcontinent and surrounding ocean regions (66.5°E–98°E longitude, 6.5°N–38.5°N latitude) at a horizontal resolution of 0.25° × 0.25°. Vertical integration from the surface up to 300 hPa across 20 pressure levels.
• Temporal Scale: Hourly resolution, covering the period from 1940 to 2024.

Methodology and Data

• Models used:
  • ERA5 reanalysis data as the base.
  • Diagnostic framework employing central finite difference methods for spatial derivatives and a backward difference scheme for temporal derivatives.
  • Vertical integration of moisture budget terms from surface pressure to 300 hPa.
  • Spherical coordinates formulation for horizontal wind divergence, including curvature correction.
• Data sources:
  • ERA5 hourly pressure-level data for specific humidity (q), zonal (u), meridional (v), and vertical (ω) wind components on 20 pressure levels (1000 hPa to 300 hPa).
  • ERA5 hourly single-level data for surface pressure (psf), vertically integrated moisture divergence (VIMD), and total column water vapor (TCWV) for construction and validation.

Main Results

• The ERA5moistIN dataset provides seven physically interpretable components of the column-integrated moisture budget: change in storage, horizontal moisture advection, horizontal wind convergence, horizontal moisture flux convergence (HMFC), vertical moisture advection, vertical wind convergence, and vertical moisture flux convergence (VMFC).
• Validation against ERA5 native diagnostics showed low Root Mean Square Error (RMSE) values (in kg m⁻² h⁻¹) across all seasons (MAM, JJAS, ON, DJF) for the period 1940–2024.
• The reconstructed total moisture convergence (HMFC + VMFC) from ERA5moistIN accurately reproduces ERA5's native vertically integrated moisture convergence (VIMC), with high Pearson correlation coefficients (r > 0.8 in most regions).
• The change in storage term from ERA5moistIN aligns closely with ERA5's time derivative of total column water vapor (dTCWV/dt), also showing strong correlations.
• ERA5moistIN reliably captures the key spatial patterns of moisture convergence during catastrophic flood events (e.g., 2005 Mumbai, 2013 Uttarakhand, 2018 Kerala, 2023 Himachal Pradesh floods), even in dynamically complex and topographically challenging regions.
• The total dataset size is approximately 318 GB, with each component stored as yearly NetCDF files (approximately 549 MB per file). Data values are in kg m⁻² s⁻¹.

Contributions

• Introduction of ERA5moistIN, the first hourly dataset of comprehensive atmospheric moisture budget components over the Indian subcontinent (1940–2024) at 0.25° spatial resolution.
• Provides high-resolution insights into moisture transport dynamics without relying on oversimplified diagnostic approximations, addressing limitations of previous studies that used coarser resolutions or simplified assumptions.
• Offers a physically consistent diagnostic framework that can be adapted to other reanalysis products (e.g., MERRA2, JRA55) and global climate model outputs (e.g., CMIP6).
• Serves as a one-stop data resource for high-resolution analysis of moisture processes, supporting studies on monsoon variability, extreme hydroclimatic event attribution, model evaluation, and predictive tool development.

Funding

• Prime Minister Research Fellowship (ID-2803578) from the Ministry of Human Resource Development (MHRD), Government of India (for A.S.R.).
• Anusandhan National Research Foundation (CRG/2023/003449) at the Indian Institute of Technology Roorkee (for A.A.).

Citation

@article{Raghuvanshi2025Hourly,
  author = {Raghuvanshi, Akash Singh and Agarwal, Ankit},
  title = {An Hourly Dataset of Moisture Budget Components Over the Indian Subcontinent (1940–2024)},
  journal = {Scientific Data},
  year = {2025},
  doi = {10.1038/s41597-025-06044-y},
  url = {https://doi.org/10.1038/s41597-025-06044-y}
}