Kumari et al. (2026) Reliability of HighResMIP CMIP6 Models for Indian Summer Monsoon Extremes and Intraseasonal Variability: Insights from Enhanced Horizontal Resolution

Identification

• Journal: Earth Systems and Environment
• Year: 2026
• Date: 2026-01-08
• Authors: Amita Kumari, Alok Kumar Mishra, Amit Kumar
• DOI: 10.1007/s41748-025-00974-8

Research Groups

• Department of Earth and Environmental Sciences, Indian Institute of Science Education and Research, Mohali, India
• K. Banerjee Centre of Atmospheric and Ocean Studies, University of Allahabad, Prayagraj, India
• Department of Civil Engineering, University of Minnesota Duluth, Duluth, MN, USA

Short Summary

This study evaluates HighResMIP CMIP6 models to assess the impact of enhanced horizontal resolution on simulating Indian Summer Monsoon (ISM) rainfall extremes and intraseasonal variability. It demonstrates that higher-resolution models significantly improve the representation of mean ISMR, extreme events, and intraseasonal dynamics, although inter-model spread and the role of physical parameterizations remain crucial.

Objective

• To examine the influence of increased horizontal resolution on persistent dry biases over India and associated physical mechanisms using HighResMIP data under the PRIMAVERA framework.
• To assess the models’ capability to capture the spatial distribution of mean rainfall, evaluate their skill in reproducing extreme precipitation indices (e.g., R95), and quantify the added value achieved through enhanced horizontal resolution.
• To diagnose the Indian Summer Monsoon (ISM) intraseasonal variability (ISV), including active and break spells and Madden-Julian Oscillation (MJO), and the monsoon low-level jet (LLJ) both qualitatively and quantitatively.

Study Configuration

• Spatial Scale: Indian subcontinent, including the Core Monsoon Zone, Western Ghats, Himalayas, Arabian Sea, and Bay of Bengal. Model atmospheric horizontal resolutions of approximately 100 km (low-resolution) and 50 km (high-resolution). Observational data at 0.25° x 0.25° (IMD) and 0.1° (MSWEP).
• Temporal Scale: Indian Summer Monsoon season (June-September, JJAS). Historical simulations. Daily rainfall data.

Methodology and Data

• Models used: HighResMIP CMIP6 models: BCC-CSM2 (MR/HR), EC-Earth3P (HR), HadGEM3-GC31 (MM/HH), MPI-ESM1-2 (NH/HR, XR), NorESM2 (LM/MM).
• Data sources:
  • Observational rainfall data: India Meteorological Department (IMD) gridded rainfall data.
  • Reanalysis wind data: ERA5 (850 hPa wind data).
  • Precipitation data for MJO analysis: Multi-Source Weighted Ensemble Precipitation (MSWEP) daily rainfall data.
  • Methods:
    • Kernel Density Estimation (KDE) for daily rainfall probability density function.
    • Expert Team on Climate Change Detection and Indices (ETCCDI): SDII (Simple Daily Intensity Index), RX5day (Maximum 5-day Precipitation), R95 (95th percentile of daily precipitation), R99 (99th percentile of daily precipitation), R95pTOT (Percentage of total precipitation from R95 events), CWD (Consecutive Wet Days), CDD (Consecutive Dry Days).
    • Taylor statistics: Root Mean Square Error (RMSE), Pattern Correlation (PC), Skill Score (SS).
    • Added Value (AV) computation: AV = 100 * (RMSEL - RMSEH) / RMSE_L.
    • Intraseasonal variability (ISV) analysis: Active and break spells (rainfall exceeding ±1 standardized anomaly for ≥3 consecutive days over the core monsoon zone).
    • Madden-Julian Oscillation (MJO) variance: 20-100 days band-pass filtered daily rainfall anomalies.
    • Low-Level Jet (LLJ) diagnosis: Relative vorticity (ζ = ∂v/∂x - ∂u/∂y) at 850 hPa, zero-vorticity line, LLJ core position and strength.

Main Results

• Higher-resolution models significantly improve the simulation of mean Indian Summer Monsoon Rainfall (ISMR) and its extremes, with rainfall distributions more consistent with observations.
• Enhanced resolution reduces dry biases over northern India and wet biases over the Western Ghats, improving spatial fidelity in regions with strong orographic influences.
• High-resolution models more realistically represent intraseasonal variability, including active and break phases of the monsoon, and show improved Madden-Julian Oscillation (MJO) variance and northeastward propagation.
• High-resolution models consistently outperform their coarser counterparts in capturing mean rainfall, extreme precipitation indices (e.g., R95, R99, RX5day), and seasonal wind patterns, particularly the Low-Level Jet (LLJ).
• The LLJ strength, position, and zero-vorticity alignment are better captured by high-resolution models, strengthening the linkage between moisture transport and rainfall distribution over India.
• Despite these advancements, considerable inter-model spread persists, emphasizing that model physics and parameterization schemes remain critical for determining simulation fidelity, alongside resolution.

Contributions

• Provides a comprehensive multi-model assessment under the HighResMIP framework, specifically for the Indian Summer Monsoon, highlighting the added value of increased horizontal resolution in capturing complex regional climate phenomena.
• Offers new insights into how enhanced resolution benefits monsoon simulations, including improved representation of mean and extreme rainfall, intraseasonal variability (active/break spells, MJO), and large-scale dynamics (LLJ).
• Quantifies performance enhancements using a diverse set of metrics (KDE, ETCCDI, Taylor statistics, Added Value) for both mean and extreme rainfall, and key dynamical features.
• Reinforces that while increased resolution is crucial for resolving fine-scale processes and reducing systematic biases, continued improvements in model physics and parameterizations are equally vital for achieving optimal skill in monsoon simulations.

Funding

This research received no external funding.

Citation

@article{Kumari2026Reliability,
  author = {Kumari, Amita and Mishra, Alok Kumar and Kumar, Amit},
  title = {Reliability of HighResMIP CMIP6 Models for Indian Summer Monsoon Extremes and Intraseasonal Variability: Insights from Enhanced Horizontal Resolution},
  journal = {Earth Systems and Environment},
  year = {2026},
  doi = {10.1007/s41748-025-00974-8},
  url = {https://doi.org/10.1007/s41748-025-00974-8}
}