The Evolution of the East Asia Summer Monsoon in June 1994 : Numerical Simulations

Alexis K.h. Lau, Anyu Wang, Ying‐Hwa Kuo, Shoujun Chen, Jimy DudhiaJournal of the Meteorological Society of Japan Ser II 1998 · DOI

Published in 1998, this research provides a detailed look at how regional climate models can dissect the complex machinery of the East Asia Summer Monsoon (EASM). By simulating the particularly active monsoon of June 1994, the study demonstrates that convective heating—the heat released when water vapor condenses into rain—is not just a byproduct of the monsoon, but the primary engine driving its evolution.

The Challenge of Modeling the Monsoon

The East Asia Summer Monsoon is a massive atmospheric system characterized by abrupt changes in wind patterns and the "sequential jump" of rain belts, such as the Mei-Yu front, from south to north. Historically, global climate models struggled with the fine-scale topography of the region, specifically the massive influence of the Tibetan Plateau.

Earlier regional models often had domains too small to capture the entire system or relied too heavily on prescribed boundary conditions. This study sought to prove that a regional model with a large enough domain could independently reproduce the monsoon's life cycle and isolate the specific physical processes, like convection, that cause the monsoon to advance.

Methodology and Experimental Design

The researchers employed the Penn State-NCAR mesoscale model (MM5), a non-hydrostatic, primitive equation model. To capture the influence of both the Indian Ocean and the Western Pacific, they used a horizontal resolution of 162 km over a massive domain covering 20°S to 60°N and 55°E to 175°E.

To understand the role of rainfall, they conducted a "Switch" experiment:

  1. CNTL (Control): A full-physics simulation including the Kain-Fritsch cumulus parameterization scheme.
  2. NCH (No Convective Heating): A version where the heat released by convection was turned off, though moisture was still removed.

The model was initialized using NCEP global analysis data from June 1, 1994, and run for 30 days. Verification was performed against NCEP wind fields, SSMI satellite-derived precipitation indices, and Chinese rain gauge data.

Model Performance and Circulation Fidelity

The CNTL simulation successfully reproduced the large-scale atmospheric features of June 1994. Specifically, it captured the northward shift of the upper-level westerlies from 35°N to 40°N and the development of the 5880-gpm contour of the Western Pacific Subtropical High (WPSH).

Comparison of 500 hPa geopotential height and wind vectors between NCEP analyses (a) and the CNTL simulation (b).

As shown in the comparison above, the model's representation of the mid-tropospheric circulation aligns closely with observed data, validating the regional model's ability to simulate the monsoon's structural evolution over a month-long period.

Convective Heating as the Primary Driver

The study's most striking insight is that the monsoon circulation effectively collapses without convective heating. In the NCH experiment, the low-level southwesterly winds—the "moisture conveyor belt"—failed to develop, and the upper-level jet stream stayed pinned south of the Tibetan Plateau.

The model demonstrated that the northward shift of the upper-level westerlies is a response to the build-up of the Tibetan High at 200 hPa, a process directly fueled by latent heat release from monsoon rains to the south. Removing this heating reduced the moisture sink over the Bay of Bengal and the Arabian Sea by approximately 150 W/m², proving that local heating is essential for maintaining the pressure gradients that pull moisture into East Asia.

Rainfall Distribution and the Mei-Yu Front

The model successfully captured the "sequential jump" of the Mei-Yu rain belt, showing that the monsoon advances through abrupt shifts triggered by interactions between tropical moisture and mid-latitude weather systems.

Comparison of monthly mean rainfall patterns: (a) SSMI satellite index, (b) observed rainfall over China, and (c) the CNTL simulation.

While the model underestimated total rainfall magnitude by about 50% in some areas—likely due to the coarse 162 km resolution—the spatial distribution was remarkably accurate. It correctly identified the three primary rainfall centers: the Arabian Sea, the Bay of Bengal, and the South China Sea.

Conclusions and Future Directions

The research confirms that convective heating is the critical driver for the northward shift of the jet stream and the intensification of the subtropical high-pressure systems defining the region's rainy season.

However, limitations remain. The 162 km grid was too coarse to resolve "meso-alpha" scale vortices that cause localized flooding. Future work should utilize nested high-resolution grids to better simulate the intense, localized rainfall events that characterize the monsoon's impact on the ground.

Summary updated

BibTeX
@article{lau1998evolution,
  author = {Alexis K.H. Lau and Anyu Wang and Ying‐Hwa Kuo and Shoujun Chen and Jimy Dudhia},
  title = {{The Evolution of the East Asia Summer Monsoon in June 1994 : Numerical Simulations}},
  journal = {Journal of the Meteorological Society of Japan Ser II},
  year = {1998},
  doi = {10.2151/jmsj1965.76.5\_749},
}
Figure