Stratocumulus clouds cover around one-fifth of the Earth’s surface and are highly reflective, sending back sunlight that could otherwise heat the Earth. They are the most favorable types of clouds from which to retrieve cloud droplet concentrations using remote sensing, but still pose challenges such as in regions of open-cell broken clouds, which are the dark regions in this image taken over the Pacific Ocean off the coast of California. Credit: Jacques Descloitres, MODIS Rapid Response Team, NASA/GSFC

It may sound like an impossible task but measuring the number of droplets in a cloud is essential for improving our understanding of cloud physics. In an article recently published in Reviews of Geophysics, Grosvenor et al. [2018] review the levels of accuracy and uncertainties with cloud droplet data gathered from satellites, and how this could improve with future spaceborne instruments. Here, one of the authors describes developments in research on cloud droplet concentration.

What are the different characteristics of clouds and how are they measured?

Clouds are complex and can be described, categorized and analyzed according to many different characteristics; it all depends on what is being studied. The main ways to measure cloud properties are aircraft and ground stations that directly sample the clouds with instruments, or via remote sensing (ground or satellite based) where the information is gathered from afar.

Our research, for example, focuses on the liquid component of clouds. From a climate change perspective, the amount of sunlight that is reflected back into space by clouds is very important, since this is energy that could otherwise heat the Earth. Liquid clouds are probably most important for this and their reflectivity is determined by their areal coverage, the amount of liquid water they contain and the number of droplets. We have been using remote sensing techniques to better understand this.

What are the strengths and weaknesses of satellite-based observations?

The great advantage of satellite observations is that large regions can be sampled (there is almost global coverage from some satellites).

Earth’s clouds as imaged from space by the MODIS satellite instrument. The image is built up from several orbits of the satellite with gaps between successive 2300 km wide swaths visible poleward of the equator. Near global coverage is achieved twice a day. Credit: NASA

This means that we can get data for regions where it is difficult to put ground based instruments, or to base aircraft flights from, such as ocean regions, very remote areas, or both!

We can also gather data over long timescales (decades sometimes), in contrast to aircraft observations that are only made a few times a day for a few weeks during field campaigns.

The weaknesses are that cloud properties are retrieved indirectly using models of, for example, how idealized clouds scatter sunlight back to the satellite instrument. Sometimes real clouds are more complicated than the modelled clouds, leading to errors.

What have been some of the recent significant scientific advances in understanding clouds?

The correct representation of clouds in climate models is vital for accurately predicting warming due to the greenhouse gases produced by humankind. Climate and weather models are getting more sophisticated and higher in resolution all the time, allowing them to model clouds more and more accurately.

The wealth of satellite observations of clouds gathered over the past 15 years or so have played a major role in helping to understand cloud processes, composition and structure so that they can be put into models and also in helping to test the models against reality, leading to model improvement.

Field campaign data (mostly from aircraft) have also been vital to this cause and are particularly useful when combined with satellite data in order to leverage the smaller scale and short duration observations to much wider regions.

Why is cloud droplet concentration a particularly useful measure?

Within a developing cloud, initially the droplet number concentration is determined only by the concentration and characteristics of aerosol particles (such as pollution particles, but also natural aerosol), and the cloud updraft, so droplet number concentration gives insight into both of these things. Other measurements that are sometimes used, such as the cloud droplet size, are affected by additional factors making analysis more difficult.

The cooling due to aerosol-cloud interactions offsets greenhouse gas warming, but this effect represents one of the largest uncertainties for climate change predictions and so further understanding is needed. Aerosol measurements within clouds are not currently possible from satellite observation and so droplet concentration provides one of the few ways to study the impact of aerosols upon clouds in the cloudy regions themselves (i.e., right where the action is happening). Cloud droplet concentration is also important for cloud processes such as rain formation, which could affect both climate predictions and weather forecasts.

Low clouds in the Northeast Atlantic off the west coasts of the British Isles, France and Portugal on 23rd December 2017 with many linear ship track features and the shipping lane visible. A visible true color MODIS satellite image (left) and cloud droplet concentration per cubic centimeter as retrieved using data from the satellite instrument (right). Credit: NASA and Daniel Grosvenor

What developments in instrumentation are needed to more accurately estimate cloud droplet concentration?

A very useful development for improved droplet concentration estimates would be a higher level of accuracy in the retrieval of cloud droplet sizes, which is an input for the technique. Much work has gone into characterizing the biases in existing instruments, but more is needed through further use of aircraft data (used as a ground truth) and high resolution cloud modelling (used as a test-bed for retrievals since it provides realistic clouds of known properties).

Newer and upcoming satellite instruments that use polarimetry are potentially a more accurate way to determine droplet size and hence concentration, but again, further validation and modelling is needed to settle questions about whether they are affected strongly by evaporation at the very tops of clouds, thus making their measurements unrepresentative. Combinations of ground-based and satellite remote sensing could complement each other and provide improved estimates, but such combinations are currently under-utilized.

—Daniel Grosvenor, National Centre for Atmospheric Science, University of Leeds, UK; email:


Grosvenor, D. (2018), How many water droplets are in a cloud?, Eos, 99, Published on 05 July 2018.

Text © 2018. The authors. CC BY-NC-ND 3.0
Except where otherwise noted, images are subject to copyright. Any reuse without express permission from the copyright owner is prohibited.