Why is sunlight hot and moonlight cool?

One measurement replaces 1000 words

This article is part of the ScienceBlogs Blog Writing Competition 2018. Information on the process can be found here. Readers can rate the articles and win a prize by voting - details can be found here. An overview of all articles participating in the competition can be found here. Information on the authors of the competition articles can be found in the respective texts.
One measurement replaces 1000 words

by Ute Parsch

I studied physics and got my diploma in astrophysics in 1992. After the birth of my son, however, I had to interrupt my professional activity. Sometimes I write and comment on the web, mostly on the subject of homeopathy, but also on other pseudoscientific topics.

My topic for this year came straight out of a discussion I had a few months ago with a flat-earth advocate. Suddenly the claim came on the table that the light of the moon had a cooling effect. After a short back and forth, the experimenter in me awoke (again) and I asked myself: What are you actually talking about against a wall instead of doing what science is all about? In addition to an hour of fresh air, the result was this article, which is now about the importance of experiments and the consideration of measurement errors and disturbance variables.

If I allow myself to have a discussion with a flat trader, then I assume that with my basic knowledge of physics and astronomy I have the “better sheet” of arguments. But it is always surprising how far this type of conversation actually takes you: As a physicist, before the conversation you don't make it clear enough what is suddenly being questioned within a few sentences, what you think Has brought argument.

This is what happened to me at the end of May of this year: Within a few sentences I was told not only that the earth is flat, but also that the moon is by no means bright because it is illuminated by the sun. Rather, the moon itself shines in a way that, in terms of its physical nature, is quite different from the light of the sun.

I admit that I did not understand the exact reason for this fact: According to the person I was talking to, it has something to do with “cold fusion of space energy” and scalar waves. Now there are not only excellent physical explanations why all the machines for using space energy do not work, as well as very good arguments against scalar waves - but also the best evidence that part of the moon is bright because it is illuminated by the sun becomes. (If we think about it for a moment, a couple of observations will come to mind: That the relative position of the sun and moon always matches the illuminated part of the moon, for example. Or that the shadows of mountains can already be seen in binoculars according to the incidence of light or that it is precisely this understanding of nature that allows us to accurately predict lunar eclipses.)

We have an interlocking web of arguments and observations for our understanding of the motions of the earth, sun, and moon. It is this interconnectedness of different observations in combination with the reliability of the conclusions confirmed in umpteen cases that gives us the right to speak of “certain knowledge” when it comes to statements such as “moonlight is reflected sunlight”.

My conversation partner from last May assured me that I could convince myself at any time of the fact that moonlight and sunlight are very different in nature - by means of a simple experiment. I would only need to use two thermometers to compare the temperature in the moon's shadow and in the moonlight on the next full moon night: I would find that the thermometer in the moonlight shows a lower temperature than that in the shadow.

Moonlight has a cooling effect because the moon - according to the flat traders - actually draws energy from the earth's atmosphere. The glow of the moon actually has something to do with this energy flow from the earth ... and nothing to do with the sun. (Please don't think of me that I find that logical in any way.) The effect would be strongest with a full moon, so it would be best to check it then. I should also take a magnifying glass and focus the moonlight a little. That way I would be able to measure temperature differences of up to eight degrees between the two thermometers.

Some background information on this claim:

In fact, this idea is relatively widespread among flat traders: moonlight cools ... and this shows us, first of all, that moonlight does not just come from the sun, but is also something very special. There are several more or less well-made videos on YouTube in which this is also demonstrated and the measurement in the moonlight actually reveals lower values: For example here or here. The author of the claim is - as is so often the case with the "flat earth" - Samuel Rowbotham. In his book “Zetetic Astronomy, Earth Not a Globe!” There is even a reference to an edition of the “Lancet Medical Journal” from 1856. Rowbotham writes that several experiments were described there that show that moonlight focused by means of a magnifying glass was used by temperature displayed on a thermometer is reduced by over eight degrees. Anyone who checks this, however, finds that by no means such experiments and results were published in the Lancet at the time, only descriptions of such impressions from hearsay.

Eight degrees. Without specifying how long I would have to cool for this and without reference to the starting temperature? But at least: eight degrees, clear work instructions ... that is a quantitative and thus actually scientifically testable prediction. In addition, an order of magnitude of the predicted effect, which I would have to see with fairly simple equipment (two refrigerator thermometers) despite the expected measurement inaccuracy if the effect is there. Nice weather and only one day until the full moon - as you can see here:

The moon on May 28, 2018, as I could use it for my measurements from my garden. The “star” next to it is Jupiter, by the way.

At this point I decided to stop arguing and just measure. Just in case anyone is surprised: No, I didn't seriously doubt that our moon would be illuminated by the sun. But that's not the point. It is the idea of ​​natural science that things are not cleared up by clever formulations and disputes, but by checking them against nature. Karl Popper defines natural science as

"... the method of making bold hypotheses and subjecting them to the sharpest criticism in order to find out where we were wrong."

Harald Lesch puts it more succinctly in "Big Bang, Universe and Life":

"Beyond good and evil, the experiment is the railing."

Very often - actually almost always - the scientific hypothesis test overwhelms the equipment that is spontaneously available in the normal household. Then we do not have the option of “trying it out for ourselves”. This is (one) reason why it is not always understood today that every scientific statement that is considered confirmed is based on a large number of measurements. And because laypeople have never looked at all of this data, many people get the impression that even scientific statements “would just be believed”.

Nobody has an LHC in the basement (and not just because of the electricity costs) or a VLT on the roof (not just because of the weight). That is why a situation like this, where you can really easily test a hypothesis, is so ideal for reconsideration. And at the same time it is a wonderful opportunity to think about a few general thoughts on the subject of “measuring”: If you are not careful, you can easily measure garbage and get results that are not very meaningful or even misleading.

For example, if we place two thermometers in the blazing sun, one on wood and one on metal, they will show different values ​​because the metal gets very hot in the sun. This also happens if we both place the two thermometers on the same material, but one on a white, the other on a black background.

If we want to carry out a scientific measurement, it is therefore important to take into account which effects could falsify the measurement results and how best to avoid such systematic errors. Even if you have thought about this beforehand, a proper measurement is part of not just throwing up your arms and shouting "Hurray" when you get a suitable result. It includes self-critically examining the question of whether there are effects other than one's own hypothesis that could have led to the same data.

This is a very important point in a scientific experiment: careful handling of the data. It is not just the fact that we conduct experiments that makes science, but also that we don't just measure “something” until it finally fits the hypothesis; we have to think about which disturbance variables could play a role, how we can reduce their influence or precisely calculate them out.

In addition to the material and the color of the subsurface, you have to look at other factors in our moonlight experiment: You have to make sure that the base and the thermometer have had enough time to come into thermal equilibrium. It is important that one of the two is not more sheltered from the wind - which can easily happen if one of the two thermometers is supposed to be in the shade. This is a conceivable source of error in some flat-ear measurements: If you simply read two different outdoor thermometers, you can easily see a certain temperature difference if the two thermometers are exposed differently, on different sides of the house and on different wall materials.

It is also important to really measure with two thermometers at the same time, because the temperature will decrease over the course of the night. So if we were to measure first in the shade and then with the same thermometer in the moonlight, the normal temperature gradient of the night would simulate the effect described by the flat traders. In addition, you have to consider the design of the thermometer: If we measure with a simple column of liquid like mine in the refrigerator thermometer, then this thermometer will only provide really good values ​​for a certain temperature range. Different thermometers with different liquids could systematically differ from each other. Because you also get a decent reading inaccuracy with such simple thermometers, depending on the angle from which you look at the thermometer, you would definitely use a different thermometer for a precession measurement, for example an infrared thermometer with a built-in thermal imaging camera, with which surface temperatures can be quickly ascertained Measure contactless. This avoids some of the problems mentioned above, but you still have to pay attention to the nature of the surface or what is in its vicinity: Such devices also give us misleading values ​​if the two measuring points that we want to compare are different Reflect thermal radiation.

Such considerations are part of every scientific measurement, not even just in physics: If, for example, we simply give cold patients a remedy and ask after 7 days whether the symptoms have gotten better, then we have an important disturbance variable in our "measurement" forget: The natural, untreated course of the disease, which often leads to feeling better after a week. Funnily enough, in "alternative medicine" such studies are repeatedly advanced as evidence of effectiveness and all anecdotal "evidence" is based on this lack of critical questioning as to which effects could have led to the observed improvements.

In order not to run into systematic errors with my own small and admittedly somewhat amateurish measurement (unfortunately I didn't have 2 expensive IR thermometers at hand), I put both thermometers as close together as possible on the same tray as a background. Before the measurement, the tray had been in the garden for a long time to ensure thermal equilibrium. The shadow for one came from a bush at a sufficiently great distance. In addition, there was no wind. I have previously tested how long the thermometers need to adjust to a new temperature and set the duration of the measurement to twice this time (it was then a total of just under an hour). I myself kept as large and above all the same distance as possible from both of them. There were no disruptive heat sources in the area.

Shortly before the start of the measurement, while trying out how best to focus the moonlight.

And then: wait. Will the doctrine of physics be confirmed? Or are the flat traders right? The voltage increases…

While we wait, we still have a little time to think about our magnifying glass: should it ultimately ensure that the moonlight thermometer is ultimately even warmer than the one in the shade? We can make a rough estimate of the incoming radiation power using the apparent brightness: the sun brings it to -26.73 mag, the moon only to -12.73 mag. Because that's a logarithmic scale, it means that the sun is about 400,000 times as bright as the full moon. Thumbs up, that means only a 400,000th of this performance on the moon. We know the radiation output of the sun, that is the solar constant and it is around 1367 W / m². We only receive a few thousandths of a watt per square meter of power from the moon.

A warming of the thermometer through my poor little magnifying glass with an area of ​​a few square centimeters is really not to be expected. And if you now run into the imagination, whether you only need a sufficiently gigantic magnifying glass to be able to ignite with moonlight, it should be said: No, no matter how big we make the magnifying glass, it won't work. The rule of thumb is that you can't make anything hotter than the surface of the original object this way ... and the moon is too cool to light anything.

I don't want to keep you tense any longer: At the beginning of my measurement, both thermometers showed 16 °. Around 50 minutes later, shortly before one o'clock in the morning, the wind came up and I finished the measurement. Both thermometers showed 15 ° there. So it cooled down by about one degree in the garden during the measurement. But that's just as much in the moon's shadow as in the moonlight.

By the way, Mike West from Metabunk has checked the assertion once more professionally than mine, condemned to improvisation: With a thermal imaging camera, not the slightest difference was found between the shadow of his car parked for hours and the asphalt that was exposed to the moonlight.

So moonlight doesn’t cool you down at all. Incidentally - but only marginally - it doesn't even have a “cooler” color temperature than sunlight. At around 4100 K, the maximum radiation of moonlight is even at longer wavelengths than sunlight. The fact that moonlight appears ashen or “bluish” is not due to the color temperature, but rather to our inability to see the colors correctly in the weak light. The phenomenon behind it is called the Purkinje effect. Just in case someone was looking for some party knowledge.

The fact that moonlight doesn't cool is actually not what is important to me about the story. I think most of them already knew that. Not even that the person I was talking to admitted that he had never carried out the measurement himself ... and has not contacted me since then.

My aim was to use a simple example to explain to you that we can sometimes save ourselves debates or tedious explanations by reflecting on the virtues of natural science and its experimental basis. This is especially a great opportunity when children ask us something. The thesis that moonlight cools could just as well have come from a child who saw two thermometers hanging in different places with different values. Then trying it out together, thinking together how to do it, how to avoid mistakes, that brings a lot more than a long (boring) explanation, from which little sticks. It is only through experiments that natural science becomes what it is.

Where this does not work due to a lack of technology, we can point out that one can often tell from the description of measurement results whether the author of a text is self-critical with his values ​​and whether he is considering what influences could have falsified the results.Where this is not the case, where measurements are designed in such a way that it is almost inevitable that they deliver “good results” (such as the follow-up observations or hand-picked individual cases in alternative medicine), we have good reason to view the measurements critically.

Real science can be recognized by the attempt to subject hypotheses to tests that are as meaningful as possible. And precisely because our secure knowledge has passed such strict tests, we can assume that we have a good sheet of arguments at hand.