GEWEX Process Evaluation Study on Warm Rain (PROES – WR)

The Global Energy and Water cycle Exchanges (GEWEX) Process Evaluation Studies (PROES) is intended to integrate observation-based metrics to understand key physical processes in climate and to improve weather and climate models at their fundamental process levels. Warm rain formation is one such process we focus on, which largely controls cloudiness and atmospheric radiative effect that affects present and future climate, and is also central to aerosol indirect effect, a major uncertainty in climate projection.

The Ultimate Goal of PROES-WR

The GEWEX PROES Warm Rain (PROES-WR) working group aims to develop new diagnostic methodologies based on observations to not only fill the gap between model performance and our knowledge of warm rain formation, but also advance our understanding the underlying physics of climate (Suzuki et al., 2018). Combining observations with model output will help us to answer questions about the current state of the Earth system and how it will change in the future.

How to Observe Warm Rain Formation Processes?

The warm rain formation process generally starts with condensation, through which the particle size tends to increase with height from cloud base. Once the particles become large enough, the coalescence process begins. In that stage, the particles fall and further collide with and collect smaller droplets lying in their path, and the particles eventually fall as drizzle or rain. Although the warm rain formation processes in clouds would have to be first monitored as rising of droplets followed by their falling from clouds as precipitation, depiction of such particle growth in vertical dimension has been an observational challenge. To overcome such challenge, a novel methodology referred to as the Contoured Frequency by Optical Depth Diagram (CFODD) was devised (Nakajima et al., 2010 and Suzuki et al., 2010). CFODD is a vertical histogram describing the probability density function of radar reflectivity normalized as a function of in-cloud optical depth (ICOD), which “fingerprints” the warm rain formation process (i.e., cloud-drizzle-rain transition) on a global scale.

 

A schematic illustration describing the condensation and coalescence processes in a warm rain cloud

 

Utility of CFODD – Model diagnostics of warm rain process

The CFODD methodology is also applicable to global model output to construct the corresponding statistics to be compared with satellite-based counterparts. Such comparisons then provide “process-oriented” model diagnostics as to how models represent the warm rain process against what is inferred from satellite observations. Prior studies have indeed demonstrated this utility of CFODD to identify biases in representing the process common among multiple global models, which can also be traced back to fundamental uncertainty in formulations of model cloud physics (Suzuki et al. 2015; Jing et al. 2017). Such “bottom-up” model constraint has also been contrasted against “top-down” requirement of model performance, such as reproducing historical climate warming, to expose their inconsistency, implying the presence of error compensations in some models (Suzuki et al. 2013a). PROES-WR will extend this approach to evaluations of up-to-date versions of global models to help improve them in representations of key aspects relevant to the warm rain process and its climatic impact as highlighted below.

Key Aspect 1: Aerosol-Cloud Interaction

Aerosol-cloud interaction, a major uncertainty in climate projection, is largely determined by the warm rain formation process that is a pathway aerosols influence cloud and precipitation. It is therefore critical to investigate how the warm rain formation is modulated by aerosols and how the modulation is represented in climate models. Early example for such an analysis with the aid of CFODD is provided by Suzuki et al. (2013b), which compared how CFODD statistics tend to vary with aerosol turbidity conditions in satellite observations and a global cloud-resolving model. Similar analysis will be extended to multiple models in PROES-WR.

Key Aspect 2: The Land-Ocean Difference in Warm Rain Formation

A recent study by Takahashi et al (2017a) used CFODD based on CloudSat/A-Train observations to show the land-ocean difference in the warm rain formation processes: the oceanic clouds had a higher fraction of tiny drizzle-size rain particles than their land-based counterparts. It was generally thought that the opposite would be true. Clouds formed with larger concentration of small drops as expected over land due to higher concentration of aerosols would tend to drizzle more and rain less heavily than clouds formed with fewer but larger drops more typical of those over oceans. So why does the opposite occur? Why clouds over the ocean are more “drizzly” than clouds over the land? The study was able to use ground-based radar data and a simple model simulation to unravel the mystery. Compared to over land, updrafts are usually weaker over ocean. Therefore, tiny particles can easily fall as drizzle over ocean. Over land, on the other hand, tiny particles cannot easily fall as drizzle because strong updrafts can force them to stay aloft. By the time they become big enough to fall, they are already rain sized. This new analysis methodology based on observations provides the first detailed analysis of the land-ocean difference in drizzle, and rain in shallow clouds and shows how understanding aerosol effects on clouds cannot be done without some understanding of dynamics.

 

CFODDs and schematic illustrations describing the warm-rain formation process over (top) weaker (i.e., ocean) and (bottom) stronger (i.e., land) updrafts.

Scope of PROES-WR

A series of observation-based analysis methodologies, including those described above, has deeply enhanced the process-level understanding of clouds. Motivated by this, the GEWEX PROES Warm Rain (PROES-WR) working group aims to develop new diagnostic methodologies based on observations. The observation-based methodologies then serve as a unified set of tools for model diagnostics when they are applied to climate model outputs and compared to the observations. The model-satellite comparisons in such manners will help improve representations of warm rain processes in models. Important topics of particular relevance to this and of climatic significance include:

  • Land-ocean comparison (e.g., Takahashi et al., 2017a)
  • Aerosol indirect effect (e.g., Suzuki et al. 2013b; Jing and Suzuki, 2018)
  • Comparison among different rain schemes and resolutions (e.g., Jing and Suzuki, 2018)
  • Exposing compensating errors (e.g., Suzuki et al. 2013a; Takahashi et al., 2017b)

The PROES-WR are now proposing multi-model comparisons exploiting CFODD and new analysis methodologies in an attempt to make direct contributions to model improvement. Modeling groups that are engaged in this effort are summarized in table below.

Institution Model
University of Tokyo/Kyushu University MIROC6
Meteorological Research Institute (MRI) MRI-ESM
University of Tokyo/NIES NICAM
Nanjing University CESM2
Geophysical Fluid Dynamics Laboratory (GFDL) AM4
The European Centre for Medium-Range Weather Forecasts (ECMWF) ECMWF
United Kingdom Met Office (UKMO) HadGEM3
University of Leipzig ECHAM
NASA’s Goddard Space Flight Center (GSFC) GEOS
Pacific Northwest National Laboratory (PNNL) DOE

Contact

Please contact Kentaroh Suzuki or Hanii Takahashi if you have any questions or comments.

References

Jing and Suzuki., 2018: https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2018GL079956

Jing et al., 2017: https://agupubs.onlinelibrary.wiley.com/doi/abs/10.1002/2017JD027310

Nakajima et al., 2010: https://journals.ametsoc.org/doi/10.1175/2010JAS3276.1

Suzuki et al., 2010: https://journals.ametsoc.org/doi/10.1175/2010JAS3463.1

Suzuki et al., 2013a: https://agupubs.onlinelibrary.wiley.com/doi/full/10.1002/grl.50874

Suzuki et al., 2013b: https://agupubs.onlinelibrary.wiley.com/doi/full/10.1002/jgrd.50043

Suzuki et al., 2015: https://journals.ametsoc.org/doi/abs/10.1175/JAS-D-14-0265.1

Suzuki et al., 2018: https://www.gewex.org/gewex-content/uploads/2018/08/Aug2018.pdf

Takahashi et al., 2017a: https://rmets.onlinelibrary.wiley.com/doi/abs/10.1002/qj.3042

Takahashi et al., 2017b: https://agupubs.onlinelibrary.wiley.com/doi/abs/10.1002/2016JD026404

 
 
https://www.gewex.org/wp-login.php?action=logout&_wpnonce=4f97cc8e72