Observations during lunar eclipses show how Earth’s atmosphere has cleared, letting in more sunlight
Strange but true: You can learn a lot about Earth’s climate by watching a lunar eclipse. This week at the 46th Global Monitoring Annual Conference (GMAC) in Boulder, CO, climate scientist Richard Keen of the University of Colorado announced new results from decades of lunar eclipse monitoring.
“Based on the color and brightness of recent eclipses, we can say that Earth’s stratosphere is as clear as it has been in decades. There are very few volcanic aerosols up there,” he explains.
This is important, climatologically, because a clear stratosphere “lets the sunshine in” to warm the Earth below.
To illustrate the effect that volcanic aerosols have on eclipses, Keen prepared a side-by-side comparison (above) of a lunar eclipse observed in 1992 after the Philippine volcano Pinatubo spewed millions of tons of gas and ash into the atmosphere vs. the latest “all-clear” eclipse in January 2018.
“Compared to the murky decades of the el Chichon and Pinatubo, the clear stratosphere since 1995 has allowed the intensity of sunlight reaching the ground to increase by about 0.6 Watts per square meter,” says Keen. “That’s equivalent to a warming of 1 or 2 tenths of a degree C (0.1 C to 0.2 C).”
“In other words,” he adds, “over the past 40 years, the decrease of volcanic aerosols and the increase of greenhouse gases have contributed equally to the total warming (~0.3 C) observed in global satellite temperature records.”
Total lunar eclipses happen somewhere on Earth typically once or twice a year. Keen is looking forward to the next one on July 27, 2018, which will be the longest lunar eclipse of the century. The Moon will pass almost directly through the middle of Earth’s shadow, remaining inside for 1 hour and 43 minutes. That’s just a few minutes shy of the theoretical maximum.
“This will give us plenty of time to measure the color and brightness of Earth’s shadow and, thus, the aerosol content of the stratosphere,” says Keen.
For more information about lunar eclipses and climate change, check out Keen’s poster from the GMAC.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
There is an eclipse here on Earth every day, as day turns into night and 12 hours later the dark emerges into the light. Anthony being a weather man should have heard this saying before ” red sky at night, sailors delight. Red sky at dawn, sailor be warned. But what does it mean Anthony ??
Mmmm … the CERES data shows a change in downwelling solar at the surface with no clouds of 0.007 W/m2 per year, or less than a W/m2 per century.
Eclipses seem like a really, really dumb way to measure the optical depth of the atmosphere … what am I missing here?
w.
Funding ?? 🙂
the data products you use have been so deeply “re-calibrated” and “recomputed” as to essentially remove signal?
https://journals.ametsoc.org/doi/pdf/10.1175/2010JTECHA1521.1
“For FM-2, the tropical mean longwave radiance during day varied by 3% over the 7-yr measurement
period in edition 1. Changes of the shortwave response of the total channel of up to 3% were made, which brought the tropical mean longwave radiance during day into agreement for edition 2. This adjustment also resulted in FM-1 and -2 radiances at nadir agreeing during daytime to a fraction of a Wm2, but the nighttime longwave differs by 1 W m2 and the shortwave differs by 2 W m2″
w. sez… Eclipses seem like a really, really dumb way to measure the optical depth of the atmosphere …
w. also sez (elsewhere) … “Please, when you comment, QUOTE THE EXACT WORDS YOU ARE DISCUSSING so that we can all understand who and what you are talking about. In addition, it’s not possible to refute someone’s claim unless you quote it first.”
so I sez … we’ve met & talked before, and I know you’re not speech impaired, so I’m at a loss to explain your “dumb” and non-specific comment.
w. sez again… what am I missing here?
so I sez… A lot, including the fundamental optics of eclipses. Read Kepler’s 400-year old treatise, extracted on the poster.
The brightness of a lunar eclipse is diminished by the passage of VISIBLE wavelength through the layer of VOLCANIC AEROSOLS in the STRATOSPHERE, GLOBALLY averaged, amplified by 40 times due to the grazing path through the layer at the limb on the Earth. So it directly measures the EXACT parameter of interest to climate studies.
I would think that it would pretty much balance out: the lack of particulates lets more sunlight in, it also lets more IR escape at night.
Considering that the rise in the average temperature of many places over the last century has been accomplished by less cooling at night instead of more warming during the day, I’d say the net effect of all the factors is not a balance between warming by day and cooling by night. And, since it is impossible to build a test chamber that will allow for convection to altitudes of 50,000 feet, there is no way to test which particles or gasses in the atmosphere are having which effect.
All we can do is check measurement records for anti-correlations. I say anti-correlations because correlation cannot show causation, while anti-correlation may show non causation.
SR
As an analogy, even when there is insufficient evidence to determine the perp in a criminal investigation, it is still possible to clear any having an alibi.
It seems to me CO2, to name one gas, has an adequate alibi to remove it from the list of possible climate perturbers
SR
The problem you don’t seem to consider is the lapse rate change at the tropopause and the cold of the stratosphere where they sulfate aerosols are located.
Think about that. The atmosphere is not a homogenous layer of gas and well mixed particles.
The moisture is preferentially confined by precipitation and cloud formation to the lower half of the troposphere. Non-condensing GHG’s are assumed to be uniformly mixed, which they probably are to the degree that matters for climate. Ozone is concentrated in the stratosphere, which is also where longer-lived volcanic aerosols were.
I have serious doubts that you can measure global average temperature (whatever that means) to 0.3 degrees C over 40 years.
Take that one up with Roy Spencer. The MSU observations are global (unlike surface station temperatures), and can measure 0.3C of warming.
If aerosols really had the forcing that climate models assume you could fix global warming in an instant.
Just turn off all the flue gas desulfurization units on coal fired power stations and use high sulfur oil in sea vessels. We’d actually save money doing so too.
Greens don’t like logical inferences like this.
The only reason aerosols are so powerful in the ensemble models is they are needed to “correct” the overegged CO2 forcing so the model output vaguely fits the 20thC verification period.
Look up the North American weather data for September 12th and 13th, 2001, when planes weren’t spewing carbon soot into the sky.
Hmm…half the warming since 1995 because of a reduction in aerosols admitting more sunshine, and it would appear that natural variation that resulted in the Dreaded Pause contributed half of the remainder at least (that was predicted by some researchers BTW, anyone with a link?). Both these contributions are based on empirical measurement, a rare commodity in climate science.
If he’s saying the the eclipsed moon was blue in 1992 and red/orange after 1995 then he’s crazy. The eclipsed moon is always red/orange. If anything an eclipsed moon after a large volcanic eruption would be more red/orange. Volcanic ash in the atmosphere filters more blue light than red light.
I’d like to see the carefully calibrated data from decades of lunar eclipses used. I bet it doesn’t exist.
Perhaps “I’m crazy, crazy for feeling so blue.”
Actually, volcanoes darken the eclipse to make them dark red, brownish, or colorless.
https://www.universetoday.com/122529/what-color-is-the-moon-a-simple-science-project-for-sunday-nights-eclipse/
Volcanic ash – particles of rock – filter all wavelengths of light equally, as do the more relevant sulfuric acid droplets (the real substance of stratospheric haze). That because ash particles and acid droplets are much larger than the wavelength of light.
Not sure what “carefully calibrated data” you dream of seeing, but the poster does show 200+ years of eclipses.
BTW, the blue is real, due to light passing through ozone in the upper stratosphere on its way to the moon. Ozone absorbs red light, and transmits blue. When volcanoes block the lower stratosphere, this upper stratosphere light gets through.
Richard Keen May 26, 2018 at 10:59 pm
I wasn’t objecting to anything particularly quotable, I was objecting to the underlying procedure itself. I’m talking about the whole idea of measuring the changes in atmospheric optical depth using eclipses.
Now, I see I didn’t specify what I thought was wrong with that idea. So let me remedy that
Here we go. For starters, you say that total eclipses occur “about once per year”. In fact, they only occur about once every 18 months. That means in thirty years we only have twenty of them … and twenty is a very small number for N.
Second, often it’s cloudy during an eclipse. So we’re down to less than twenty for N in 30 years.
Third, there are a variety of other things that affect the atmospheric optical depth through which you are taking the photograph. Smoke, haze, temporary aerosols, ice crystals, very thin clouds, small salt crystals, and water vapor all affect the AOD. How can you remove these from a one-off observation? Yes, you can compare the moon to nearby stars, but the colors of the stars and the moon are quite different, so the losses will be different—atmospheric absorption of light is frequency-dependent.
Fourth, each time the observation is done from a different spot on the planet … which will likely be at a different altitude, with different permanent, semi-permanent, and temporary conditions affecting the conditions of the photograph.
Fifth, if the photo is not taken at the exact instant of totality, the calculations for the theoretical brightness will e off by some unknown amount.
Mmmm … first, you are not averaging just the volcanic aerosols in the stratosphere. The light hitting the moon doesn’t just go through the stratosphere. It passes through the troposphere as well. You are using an “assumed cloud distribution of 50% in the troposphere”, but obviously nature is never that regular. The light is passing through a thin band at the terminator, and there is no reason to assume that clouds in that thin band, which is located at a different part of the earth for each eclipse, is either consistent or 50% … and many reasons to believe it is not consistent or 50%.
Next, the temporal resolution of the eclipse data is quite bad. You have carefully picked six volcanoes in your poster. As you describe it, “Likely VEI = 4 or 5 Noted”.
But that is only a small fraction of the VEI 4 – 5 eruptions during the time period. I digitized your AOD data for the post-1995 part of your poster, and added the VEI 4-5 eruptions during that time period.
I’m sorry, but I’m just not seeing the correlation between the AOD (orange circles) and the VEI and timing of the eruptions (gray dots). I also don’t understand why the orange line has bends in it when there were no eclipses during that time.
So … those are my objections to the procedure. My apologies for my lack of details before. I was moving fast, looked at it, saw the whole host of problems I listed above, and went “whaaa?
Thanks for calling me on it in a pleasant fashion, my best to you,
w.