A Global Model Investigation of MJO Initiation for DYNAMO

The DYNAMO field campaign in the Indian Ocean provides an unprecedented opportunity to study MJO initiation. In this project, the PI proposes to investigate the MJO initiation in the Indian Ocean using the NCAR CAM3 and the DYNAMO observations. The objective is to improve MJO simulation, and ultimately MJO prediction using global models. As part of the DYNAMO modeling effort, the project aims to answer the following scientific questions relevant to Hypotheses I and II in the DYNAMO Science Planning Overview (SPO) document:

I. What are the factors determining the initiation of MJO in the Indian Ocean?
II. How does the cloud population interact with the MJO circulation during MJO initiation? Can the NCAR CAM3 reproduce the observed cloud population?

The basic research tools used in this work are the NCAR CAM3 and the improved Zhang-McFarlane convection scheme. The data used for model evaluation and improvement will be DYNAMO field observations from sounding array, radar and satellites, ECMWF reanalyses products, and other data assembled by the Year of Tropical Convection (YOTC) project. The combination of model and observations will allow us to test new ideas using models and evaluate them using observations. To answer the above questions, we will perform a series of simulations using the CAM3 and its single column version in both prediction mode and the traditional climate simulation mode. Simulation tests will be conducted to determine what factors affect the MJO initiation the most. In particular, shallow convection preconditioning, convective sensitivity to environmental moisture through lateral entrainment, sea surface temperature and surface evaporation, among others, are deemed critical to MJO initiation. We will investigate each of them in the proposed research. The model output will be compared with DYNAMO observations and reanalyses products. For each observed and simulated MJO, budgets of heat and moisture will be computed to determine the sources and sinks of moisture during different stages of MJO initiation over the Indian Ocean. The cloud population, as measured by cloud top heights and optical depth, during the MJO life cycle from model simulations and satellite observations will be compared and related to MJO circulation. We will also participate in the inter-model comparisons to identify sensitivity of different models to parameters in convection scheme.

Intellectual merit: MJO simulation is a challenging scientific problem in GCMs. Using an improved version of the convection scheme developed by the PI, the NCAR CAM3 can simulate MJOs realistically. Thus, the model can be used as a tool together with DYNAMO observations to understand MJO initiation in the Indian Ocean. The work will help improve the MJO simulation and prediction not only in CAM3, but also in other GCMs.

Broader impacts: Poor MJO simulation is a well recognized problem in many GCMs. It negatively impacts the model development efforts in the global modeling community, and affects the simulation of other climate systems. This work will have impacts beyond the MJO dynamics and simulation. MJO simulation can be used as a metric to evaluate model performance.