Studying convective and cloud processes during the MJO and evaluating their representation in climate models by combining humidity, cloud and water isotopic measurements <br />
The Madden-Julian Oscillation (MJO) is the dominant mode of intraseasonal variability in the tropical atmosphere. However, climate <br />models have persistent difficulties in simulating its characteristics. Why do some models capture the MJO better than others? What processes are key to simulate the MJO? In this proposal we try to address these questions by combining humidity and cloud measurements with water vapor isotopic measurements (HDO/H2O ratio).
(1) What controls the capacity of the model to represent the large scale organization of convection and its role on its environment?
-> Compare sensitivity tests to the model physics in 3D with nudged winds or in 1D with imposed vertical velocity.
-> Understand the sensitivity.
-> Compare simulations with A-train/TES and IASI observations.
-> Establish a framework to interpret isotopic observations.
-> Deduce possible model biases in their simulation of the role of convection on environmental moisture.
-> Suggest how improvements in the model physics could reduce these biases.
(2) What processes make the MJO specific compared to other modes of intra-seasonal variations?
-> Compare composites of MJO events and of other types of events.
-> Compare simulations and observations.
(3) What are the relative roles of the representation of different convective processes and radiative/dynamical feedbacks in explaining model biases in the MJO simulation?
-> Compare nudged and free 3D simulations
-> Compare 1D simulations with imposed or interactive vertical velocity.
-> Compare sensitivity tests.
-> Understand how physical biases identified in (1) impact the large-scale dynamics and MJO simulation.
- Precipitation deletes the water vapor more when it's related to stratiform processes than when it's related to convective processes. The convective/stratiform proportion determines the sensitivity of precipitation to tropospheric humidity
- Precipitation deletes the water vapor more when it's associated with deep mixing and a top-heavy vertical profile of large-scale vertical velocity.
- The joint evolutions of humidity and isotopic composition is related to the relative timing of the different convective and cloud processes during MJO events.
- publish papers
- understand the link between the capacity of a model to simulate convective processes for a prescribed large-scale dynamics, and its capacity to simulate the intra-seasonal dynamics in a free mode.
- look at the impact of convective organization.
O. Tuinenburg, C. Risi, J.L. Lacour, M. Schneider, A. Wiegele, J. Worden, N. Kurita, J.P. Duvel, N. Deutscher, S. Bony, P.F. Coheur and C. Clerbaux. Moist processes during MJO events as diagnosed from water isotopic measurements from the IASI satellite. S
The Madden-Julian Oscillation (MJO) is the dominant mode of intraseasonal variability in the tropical atmosphere. However, for several decades however climate models have met difficulties in simulating its properties. The development and propagation of the MJO involves the interaction between convective, cloud, radiative and dynamical processes, whose representation in climate models bears uncertainties. The same processes happen to be crucial also in climate change projections.
The goal of the proposal is to understand why some models capture the MJO better than others. What processes are key to simulate the MJO? These questions are not new, but the novelty of this proposal is to address these questions by combining humidity and cloud measurements with water vapor isotopic measurements. Several studies have shown the added values of water isotopic measurements to study convective processes. More specifically for this MJO, a recent study has shown that the water stable isotopic evolution during the MJO provides a complementary information compared to the evolution of meteorological variables.
In this project we will blend and analyze three datasets: (1) the A-train satellite dataset, combining TES isotopic data and Calipso and Cloudsat cloud data, with the vertical resolution as a main advantage, (2) satellite, measuring both isotopic and cloud properties from satellite, with the spatio-temporal coverage as a main advantage, and (3) the isotopic, meteorlogical and cloud measurements at the Darwin ARM site, with the temporal resolution as a main advantage.
These datasets will be used to evaluate isotopic simulations with the LMDZ iso general circulation model (GCM) LMDZ. Sensitivity tests to convective and cloud processes will be compared to identified critical processes in the MJO simulation. The use of LMDZ is particularly adequate and timely in this context: the new version of the LMDZ GCM includes an improved representation of convective processes and exhibits a more realistic MJO variability compared to the previous version. Comparing with an isotopic version of the cloud resolving model Meso-NH will allow us to evaluated convective processes in more detail. Different modele configurations with LMDZ will be compared to quantify the relative effects of physical biases and dynamical feedbakcs. LMDZ will be compared with other isotopic GCMs to check that the representativity of our results in the context of climate models diversity.
The expected outcomes of this project are to :
(1) design a framework to interpret joint humidity, isotopic and cloud distributions in terms of convective processes ;
(2) identify the critical processes to simulate properly the MJO and which of these are commonly mis-represented in GCMs ;
(3) suggest parameterization improvement.
Madame Camille RISI (Laboratoire de Météorologie Dynamique) – firstname.lastname@example.org
The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.
LMD Laboratoire de Météorologie Dynamique
CNRS DR ILE DE FRANCE SUD
Help of the ANR 147,000 euros
Beginning and duration of the scientific project: January 2013 - 48 Months