International Joint Project to Study Warm and Cold Processes of Precipitation
Project leader: Dr V. Phillips, University of Lund, Sweden
Co-Investigator: Professor J. Martins, Department of Physics, Federal University of Technology, Londrina, PR, Brazil
Co-Investigator: Professor Fabio Goncalves, Department of Atmospheric Sciences - Institute Astronomical and Geophysical, University of Sao Paulo, Sao Paulo, Brazil
Overview
A project entitled “Comparison of Warm Rain and Ice-Crystal Processes for Precipitation and Lightning in Cold Clouds” led by Lund University was funded in summer 2018 by the Crafoord Foundation in Sweden. It lasted a couple of years (2019-2021), supporting a postdoctoral scholar for 5 months.
This joint project involved three institutions:
· Department of Atmospheric Sciences - Institute Astronômical and Geophysical, University of Sao Paulo in Brazil;
· Department of Physics, Federal University of Technology, Londrina in Brazil;
· Department of Physical Geography and Ecosystem Science, Lund University, in Sweden.
The aim was to predict the separate contributions of precipitation from two processes in our existing high-resolution model of clouds, described below. The two processes are:
· coalescence of cloud-droplets to form rain (‘warm rain process’), which may freeze at subzero levels or may fall out as ‘warm’ precipitation;
· and growth of ice crystals to form snow (‘ice crystal process’), which may rime to form graupel/hail or may fall out as ‘cold’ precipitation.
There is much uncertainty about their relative activities because both processes can co-exist in any given deep cloud (e.g. ‘mixed-phase clouds’ consisting of ice and liquid). Also, ice is complex in its morphology and inherently challenging to simulate, even approximately, with numerical models. Note that surface rainfall may be either ‘warm’ or ‘cold’ or a mixture of both. However, snow can only form by the ice crystal process and is always ‘cold’.
The terminology, ’warm’ or ‘cold’, refers only to the mechanism of origin of the precipitation. It implies nothing about the current temperature of the precipitation or its current phase, whether that be ice or liquid.
Mixed-phase clouds promote the ice crystal process because their co-existence of ice and supercooled cloud-liquid causes the humidity to be close to water saturation. This in turn causes the humidity to be well above saturation with respect to ice, so that any ice crystals can be nucleated by ‘ice nucleus’ (IN) aerosols and grow fast by diffusion of vapour to sizes at which they aggregate to form snowflakes.
Mixed-phase clouds require sufficient ascent to maintain the humidity near water saturation so that the supercooled cloud-liquid can grow by condensation somewhere in-cloud. However, when the ascent is weak yet sufficient for mixed-phase conditions, the humidity can be maintained just below water saturation and the supercooled cloud-liquid slowly evaporates while ice crystals grow. This special type of ice crystal process is termed the ‘Wegener-Bergeron-Findeisen’ process.
Generally, the cloud-base temperature in conjunction with local aerosol conditions of the environment controls whether the cloud-droplets in any cloud are large enough to collide, and hence if the warm rain process is active. The cloud-top must be at subzero levels for any ice to form from the activity of ice nucleus (IN) aerosols in cloudy ascent, so that the ice crystal process can occur.
Thus, the environment and cloud-type determines the balance between warm and cold precipitation in any cloud in a multi-facetted way. While the linkage between environment and precipitation mechanism is qualitatively understood, it is unclear to what extent the warm or cold mechanism prevails in various cloud-types. Hitherto, the quantitative details have remained elusively uncertain.
It is this linkage that the project elucidated for selected cloud-types from observed cases.
Schematic pictures of warm rain process (left) and ice crystal process (right) for precipitation production.
Novel approach to analysis of numerical simulations
There were two aspects of the approach.
First, ‘tagging tracers’ were applied in the model so as to follow components of the rain, snow and graupel/hail concentrations that are due to both processes. The tagging tracers are passive, having no effect on the simulation. The total rainfall at the surface would have components from warm and cold rain that would be predicted in the same simulation.
This was a superior approach compared to the alternative. The alternative approach would be to switch off artificially either warm or cold process to predict the change in precipitation. That would be problematic since the clouds would be radically altered and non-target processes would be affected by compensation. We would then be studying alien clouds instead of the actual clouds in the investigation.
Second, several contrasting observed cases were simulated with validation of each simulation against coincident aircraft, ground-based and satellite measurements. This diversity of clouds created a stringent test of the model, creating confidence that the model was depicting natural clouds. The ice concentrations were shown to be predicted with adequate accuracy in all cases. The wide span of contrasting cases enabled more fundamental conclusions to be drawn about the factors controlling the balance of warm-vs-cold precipitation.
Observed cases of clouds simulated
Three cases with contrasting cloud-base temperatures were simulated, with comparison of each simulation against coincident observations:
· GO-AMAZON (funded by DoE): Very warm based convective systems in Brazil were sampled by aircraft (G-1), and aerosol conditions outside clouds were measured;
· MC3E (funded by DoE/NASA): Slightly warm-based clouds in precipitating deep convective systems were sampled by aircraft during Spring 2011 over Oklahoma, USA;
· STEPS (funded by US National Science Foundation [NSF]): Cold-based clouds in deep convective systems were sampled by an armoured aircraft that flew through the fast cloudy updrafts during summer 2000 over the US High Plains in Colorado/Kansas.
Description of cloud model
The aerosol-cloud model (AC) predicts microphysical and dynamical properties of clouds. It resolves clouds typically on a grid with a mesoscale domain (e.g. 100 km wide) and a resolution of 1 or 2 km. AC represents 7 chemical species of aerosol: primary biological aerosols, non-biological insoluble organics, soluble organics, sea-salt, ammonium sulphate, mineral dust and soot. Interstitial and immersed/embedded components of each aerosol species are predicted in cloud and precipitation. The model predicts active ice nuclei (IN) and cloud condensation nuclei (CCN) concentrations from the chemistry, sizes and loading of aerosols. There are 5 microphysical species: cloud-droplets, ice crystals, snow, rain and graupel/hail.
Warm-vs-cold precipitation: model results
In all three cases, adequate agreement, for many diverse cloud properties and precipitation rates, between each simulation and coincident observations was obtained. The AC model was comprehensively validated.
Analysis of the tagging tracers showed that both cloud-base temperature and CCN aerosol conditions controlled the relative abundance of cold and warm rain. The Brazilian storm (Go-AMAZON) with a warm base (> 20 oC) developed predominantly warm rain at the ground, while the storm over the High Plains (STEPS) with a cold base (near 0 oC) developed mainly cold precipitation. The evolution of accumulated surface precipitation (plots below) shows this is true both for the entire storm and for the convective and stratiform regions individually. In the Brazilian case, the rain developed by coalescence during convective ascent before the freezing level was reached.
Moreover, sensitivity tests showed that lowering the soluble aerosol loading in the environment could promote warm rain slightly. Also, lowering the cloud-base with updraft speed statistics held constant also boosted warm rain.
Accumulated surface precipitation for the STEPS case of cold-based convective clouds on 19/20 July 2000 over Kansas/Colorado in USA. From Gupta et al. (2023).
Accumulated surface precipitation for GoAmazon case of very warm-based clouds on 19 March 2014 over central Brazil near Manaus. From Gupta et al. (2023).
Moreover, the aircraft during the STEPS storm (19/20 June 2000) observed an absence of any supercooled raindrops aloft in the convective cloud system. This lack of a warm rain process is reproduced by the AC model (see plot below). There is no formation of graupel/hail by raindrop-freezing, since no warm rain can form, and instead the graupel/hail is predominantly ‘cold’.
More details of the results are shown in a paper currently in review (Gupta et al. 2023).
Current progress in the project
The project was completed according to plan. The tagging tracers were added to the cloud model (AC) and a variety of contrasting cases allowed the warm-vs-cold precipitation balance to be quantified. The amounts of ‘warm’ and ‘cold’ particles of graupel/hail and rain were predicted in vertical profiles through the simulated storms.
A paper describing results was submitted to a journal in 2022 (Gupta et al. 2023). Also, at the International Conference on Clouds and Precipitation (ICCP) in 2022 organised by Pune, India, a talk was given by the postdoctoral scholar about the results. Thus, although the postdoctoral scholar was only at Lund for the duration of project funding of about half a year in 2019, subsequent collaboration continued informally yielding a comprehensive set of results.
Bibliography
Gupta, A. K., Waman, D., Deshmukh, A., Jadav, A., Patade, S., Phillips, V. T. J., Bansemer, A., Martins, J. A., and F. Goncalves, 2023: The microphysics of the warm-rain and ice crystal processes of precipitation in simulated continental convective storms. Nature Comm., in review