# Entropy and evolution controversy

1. Jan 11, 2010

### raimo lonka

Recently several papers have been published related to entropy and evolution. According to Daniel F. Styer ("Entropy and evolution", Am. J. Phys 76, Nov 2008) and Emory F. Bunn ("Evolution and the second law of thermodynamics" Am. J. Phys 77, Oct 2009) entropy is decreasing due to evolution. Styer gives an estimate for the entropy change due to evolution, which is - 302 J/K per second. According to Styer and Bunn the decrease in entropy due to evolution is not in conflict with the 2nd law of thermodynamics because Earth is "constantly absorbing sunlight resulting in an enormous increase in entropy." Bunn writes: "...the rate of entropy increase due to the Earth's absorption of sunlight must be sufficient to account for the rate of entropy decrease required for the evolution of life (a negative quantity)."

According to Annila et al. [See eg. Vivek Sharma and Arto Annila, "Natural process - Natural selection," Biophys. Chem. 127, 123-128 (2007) ; Ville Kaila and Arto Annila, "Natural selection for least action," Proc. R. Soc. A 464, 3055-3070 (2008); Salla Jaakkola, Sedeer El-Showk and Arto Annila, "The driving force behind genomic diversity," Biophys. Chem. 134, 232-238 (2008)] entropy is increasing during evolution. Annila et al. writes: "There is no need to explain the rise of orderly structures by invoking an exemption that entropy would decrease in a living system at the expense of its surroundings. Entropy is increasing in living systems as well by dispersal of energy." "The primitive chemical evolution took the direction dS/dt > 0, just as the sophisticated evolution does today." "It is possible to deduce the direction of evolution and ensuing overall distribution of the genomic entities by requiring that S will increase until dS/dt = 0." "The rate of entropy increase can be regarded as a universal fitness criterion of natural selection."

So, we have peer-reviewed papers that are making totally opposite claims. According to Styer&Bunn entropy is decreasing (a negative change) due to evolution and according to Annila et al. entropy is increasing (a positive change) due to evolution. The negative change of entropy is also an important part of the argument presented by Styer&Bunn, because they use the increase in entropy due to Earth's absorption of sunlight to legitimate the claim that there is no conflict between evolution and the 2nd law of thermodynamics. And for Annila et al. entropy increase is also an important part of the argument: “...entropy increase can be regarded as a universal fitness criterion of natural selection”.

How is this controversy possible? We have a situation equal to a situation where some peer-reviewed papers would claim that gravity is pulling masses together while some other peer-reviewed papers would claim that gravity is pushing masses apart. What kind of science is this? Should we believe in Styer & Bunn and say that evolution is decreasing entropy or should we believe in Annila et. al and say that evolution is increasing entropy?

2. Jan 11, 2010

### qedprigmosyno

Hi,
The second paper looks interesting and I am going to go read it now.
To answer your concern, it looks like what is being described is the energy cycle, that is if you intertwine the two articles. One addresses obtaining energy and it the other one seems to address the dispersion of energy. I think what was concerning was that the two articles had seemingly opposing views of entropy as a result of a live dynamic world; you must understand that entropy is always changing in any environment. Sure from a molecular level entropy is reduced to form complex molecules, but entropy is also increases at the anatomical level to maintain homeostasis and other functions. Thus, it is important to realize at what aspect are the scientific papers looking at, otherwise, you'll lose the grand scheme.
To address the secondary concern, science is a procedure to model the natural world; this is regardless of the stances they take. If there are two theories, one stating gravity is pulling apart matter and one stating gravity is pulling matter together, well then we resort to experiments to demonstrate that one of these theories is correct to our everyday understanding of nature; at this point, they are only hypothesis, or just guesses. What is key to science is the utilization of evidence and experimentation.
I hope this helps, if not, post more questions and i'll see that they get answered.

3. Jan 11, 2010

### sylas

Your two papers are saying the same thing in different ways.

None of this is new research. In particular, the article by Styer is intended to be used to help high school students and undergraduates understand something that has always been clear from when evolution was first defined.

Forget evolution, and think instead about biological growth. You plant a seed, and you end up with a tree. The material that makes up the tree is more organized than the seed and the nutrients provided to let it grow. Does this violate thermodynamics? Of course not.

The second paper explains this by pointing out that the total change in entropy is an increase, when you take into account the useful energy (especially sunlight) which powers the whole process and is then given back as waste energy (especially thermal).

The first paper does the same thing. The first paper does it by walking through a simple calculation of the entropy change in the tree itself (or rather, an upper bound) and then compares that with the global change. Or rather, the first paper walks through the calculation using evolution rather than growing a tree, but the principle is precisely the same. The local organization of a tree, whether you consider how it grew from a seed or evolved from simpler ancestors, is a physical process consistent with the second law, which overall produces a heck of a lot of entropy (waste energy) while at the same time giving an itsy bitsy little bit of useful local organization (a tree).

-------

Here is a reproduction of the calculation of 302 J/K/sec. I am indebted to PZ Myers at the Panda's Thumb for details, since I could not get the original paper myself, but I have fixed a problem with mixing up centuries and years.

Note that this is a very generous upper bound; not a precise value. We often use upper bounds like this to prove consistency with thermodynamic laws.

First, Styer proposes that each individual is no more than 1000 times "more improbable" than their ancestors from one century ago. That is, that we are specified a thousand times more precisely than our great-grandparents. This clearly much much more than is needed; the idea is to give the most generous possible upper bound. This can be represented:
$$\Omega_f > 10^{-3} \Omega_i$$​
where Ωi and Ωf are the number of "microstates" for the initial and final organism. Using statistical entropy definitions, we have entropy S as
$$S = k_B \ln \Omega$$​
where kB = 1.38*10-23 J/K is the Boltzmann constant. A century is 3.156*109 seconds, and so the rate of entropy change getting from ancestral to modern organisms is
\begin{align*} S' & < (S_f - S_i) / 3.156\times 10^9 \\ & = 1.38\times 10^{-23} ( \ln \Omega_f - \ln \Omega_i) / 3.156\times 10^9 \\ & = 4.37\times 10^{-33} \ln 10^{-3} \\ & = -3.02\times 10^{-32} \; \text{J/K/sec} \end{align*}​
That's for a single organism. So how many organisms are there altogether on the Earth? To a first approximation, everything is a prokaryote (bacteria), and 1030 was the estimate he used. End result... -302 J/K/sec

But of course, that isn't all that is going on. Styer also calculates the total entropy changes going on at the same time as living processes are being driven ultimately by sunlight. Earth absorbs about 240 W/m2 as sunlight, with a temperature of about 5780K. It radiates it back again with a temperature of about 255K. So the rate of entropy gain per second is
$$240 \times 5.1\times 10^{11} \times (1/255 - 1/5780) = 459 \times 10^{12} \; J/K/sec$$​

Added in edit. A better calculation might be to use 288K for the temperature of the Earth, which is at the surface. Most waste heat from biological processes will be at this kind of temperature rather than at the effective emission temperature. This would give 404*1012 J/K/sec.​

So the gain in entropy from Earth's processing of sunlight is about 12 orders of magnitude more than what would be needed to drive evolution.

Not everyone agrees that this is a useful teaching approach, but if it helps someone understand why the entropy related objections to evolution are incorrect then it has been useful.

Remember, this is not advanced research we are talking about. It's simple calculations to help students with common misconceptions about how thermodynamics relates to evolution.

Cheers -- sylas

Last edited: Jan 12, 2010
4. Jan 11, 2010

### Staff: Mentor

Sounds like the difference is that one quantifies the entropy generated by living things when they burn energy whereas the other only quantifies the entropy reduction/storage by the growth of their bodies. It doesn't subtract out the entropy lost in burning the energy to do that.

5. Jan 12, 2010

### baywax

What I'd like to know, kind of on topic here, is where heat "goes" and how can entropy take place when energy cannot be destroyed?

6. Jan 12, 2010

### sylas

Entropy has units of J/K, which is energy per unit temperature.

Roughly, the same amount of energy will have different entropy at different temperatures, so although energy is conserved, entropy is not.

For example. Suppose you have one object with a temperature of 5780 degrees, and another at 255 degrees. Suppose you have 1000 J of energy flowing from the the hot object to the cooler one.

The entropy difference is 1000/255 - 1000/5780, which is about 3.75 J/K. This is the entropy increase.

Those numbers correspond to the temperatures of the Sun, and the "effective radiating temperature" of the Earth. Energy flowing from the Sun to the Earth gives an increase in entropy.

Note, however, that the Earth radiates that energy away once more. So where does it go? Basically, it goes out into empty space, which has a temperature of about 2.725K (The background radiation.) So the entropy change as Earth radiates waste heat into space is about 1000/255 - 1000/2.725 = 363 J/K. This is another entropy increase.

The Sun is not a closed system. It is decreasing in entropy; but that is consistent with the laws of thermodynamics, because the loss of radiation out into space represents a much larger gain in entropy over all. Earth sits in the middle of this flow of energy from the Sun. It gets "hot energy" from the Sun, and then radiates the same amount into frigid space. In the meantime, all kinds of processes can tap into this energy flow and do useful works. Waterfalls, weather, rain, waves, ocean currents, living things, solar panels; they all work because this entropy difference means the energy can be used to do work.

By the way, it is a common error to think that because processes on Earth are driven by sunlight, the Sun must have the gain in entropy to satisfy the second law. Not so! The gain in entropy is always at a cooler body, and in the heat sink where waste heat is dumped. For processes on Earth, the place where heat is dumped is empty space. That ultimately takes up the gain in entropy to satisfy the second law for those processes on Earth which involve small pockets of organization and entropy decrease.

Cheers -- sylas

Last edited: Jan 12, 2010
7. Jan 12, 2010

### Andy Resnick

This is an amazing post- mind if I (re-)use it in class?

8. Jan 12, 2010

### Andy Resnick

The last paper (Biophys. Chem. 134) was removed (although still available on line, as is the retraction note). Looks like I have some reading to do today...

9. Jan 12, 2010

### sylas

By all means. I personally tend to take a slightly different approach, but what works for different students will vary.

I would also encourage you to go to the original source. (Ironically, however, I didn't.) Unfortunately, it is behind a paywall, but you may have ways around that. You could certainly email Professor Daniel Styer. The reference is:
• D.F. Styer, (2008) http://dx.doi.org/10.1119/1.2973046" [Broken] in Am. J. Phys. 76, 1031-1033. Reprinted in the Virtual Journal of Biological Physics Research 16 (15 October 2008).

Note that the http://www.vjbio.org/ajp/ [Broken], but I believe that article has an error in the maths, which I fixed in my post above. PZ Myers mixed up centuries and years inconsistently. Also, he gives 420*1012 J/K/sec for the entropy change from sunlight; but without a calculation. So I substituted one of my own. The order of magnitude is the same.

Cheers -- sylas

Last edited by a moderator: May 4, 2017
10. Jan 12, 2010

### raimo lonka

In my starting post I mentioned only 3 papers (sry, my laziness) written by Annila et al., but actually there are 5 of them addressing the entropy and evolution topic. Others are: Mahesh Karnani and Arto Annila, "Gaia again," Biosystems 95, 82-87 (2009) and Vivek Sharma and Arto Annila, "Natural process - Natural selection," Biophys. Chem. 127, 123-128 (2007).

In "Natural process - Natural selection" Annila et al. write: "Entropy will increase when increasing numbers of compounds emerge from reactions...", "Entropy will also increase when various kind of compounds emerge from syntheses...", "Entropy will also increase when the system acquires more compounds to its processes from surroundings."
"Entropy will increase when simple compounds, i.e. low-j entities react to form more complex compounds..."
"Therefore, we propose that the rate of entropy production is the fitness criterion of natural selection".

So it seems that Annila et al. do not agree with you on a molecular level and I still maintain that there is a conflict with Styer & Bunn.

regards,

raimo

P.S. Professor Annila (arto.annila at helsinki.fi) kindly sent me all the papers when I asked for them.

11. Jan 12, 2010

### sylas

There's one basic fact of life that applies here: the second law of thermodynamics. The thread title suggests that there's some controversy involved here.

Truly, there is not; simply a case of understanding what the papers are talking about.

Every physical process always results in a net increase in entropy. There's nothing in any of the research cited here, or in any other legitimate research, that proposes anything in evolution to conflict with that: no matter whether you look at molecular levels or at metabolism levels or at planetary levels. Any level you pick; any aspect of evolution you pick; and there is a net increase in entropy involved.

If you think you have found a paper that disagrees with that (in this thread, none of the papers disagree with it) then you either don't understand the paper, or else you have found a crackpot paper, on a par with perpetual motion machine "papers".

The statements you list as premises are unexceptional.

The conclusion doesn't actually follow from those premises as worded, but the proposal of identifying rates of entropy production with evolutionary fitness is not unique to Annila and has been around for quite a while. The late Iyla Prigogine received a Nobel prize for work on non-equilibrium thermodynamics and dissipative systems; and he also had similar ideas. So let's skip over fitness criteria and simply note that Annila's remarks on entropy are unexceptional consequences of what we all know... EVERY process works to increase entropy. He's going into the details of those processes in an interesting way, but let's clear up the basic stuff first!

I have not entirely understood what qedprigmosyno is saying. This remark seems to miss the point:
I don't think "levels" come into it this way. From a molecular level, entropy is INCREASED to form complex molecules. Entropy increases with every process, no matter what "level" you look at. You just have to make sure you give a complete list of the changes that are involved. Also -- be cautious of presuming a complex molecule has less entropy than the substances it was formed from. Complexity actually has very little to do with it.

But as for the alleged conflict with Styer and Bunn, you seem to be ignoring explanations and simply holding on to a misconception no matter what is explained.

Styer and Bunn both explicitly note that entropy increases as well. All they are doing is putting bounds on any possible localized decrease in entropy involved, and comparing that with the vastly larger increase in entropy in the associated energy flows that are used to drive all life processes. The paper by Bunn is a little bit more sophisticated, and can be freely downloaded from Evolution and the second law of thermodynamics at arXiv:0903.4603v1. It is proposed as a more robust formulation of Styer's simple calculation.

There is no conflict involved between these papers. They all work in the recognition that evolutionary processes are increasing entropy. There can be localized decreases in entropy which can easily be given a strong upper bound -- and that is all Styer and Bunn use, an upper bound on localized entropy reduction.

Annila is much detailed, and in particular this work can help address another common confusion, which is thinking that there is necessarily some significant local reduction in entropy involved in organized structures.

But since Bunn and Styer are only giving generous upper bounds on any local entropy decreases while also recognizing associated much larger increases, there's no conflict with Annila's more sophisticated examination of the actual processes involved in living things. Annila proposes that living things are finely organized structures that are really good at processing energy in useful ways -- and that a fitter organism is often better at processing and using more of the energy. This means that evolution is producing structures that are dissipating energy faster than ever. They are hence also better and better at raising entropy. (Dissipative systems.)

For readers looking on who want to read Annila's work for themselves, try Natural selection for least action, by V.R.I. Kaila and A. Annila, in Proc. R. Soc. A (2008) 464, 3055–3070, doi:10.1098/rspa.2008.0178, free access.

Cheers -- sylas

12. Jan 12, 2010

### baywax

Whoa! thank you sylas... great in depth answer to my ignorance!

So the heat is lost but it is still doing work in some form or format... it cannot be destroyed. If we just say it goes into empty space this could imply it is dissipating into the general background temperature of 2.725K and contributing to that radiation event.

13. Jan 13, 2010

### sylas

It is a bit tricky dealing in detail with the radiation into space. I don't know what you mean by a "radiation event". You can't really give a temperature to a photon; but you can give a temperature to a "bath" of radiation with the Planck spectrum.

Space is filled with radiation, of all kinds; and it is not in thermal equilibrium. So you can't really say it has a single temperature, just as you cannot give a single temperature for a metal rod that is hot at one end. You have to wait until it is in equilibrium before it has a well defined temperature. If you extend your line of sight in most directions indefinitely, you end up looking at the "surface of last scattering" which looks like a blackbody surface covering all the sky, and which has a temperature of 2.725K. It is because this is so cold that the sky looks dark. For our purposes here, the thermal emissions from Earth work a bit like emission into a sink at that temperature. The point is that thermodynamics is satisfied because Earth can easily shed waste heat into empty space.

Cheers -- sylas

14. Jan 16, 2010

### edpell

No. If you focus on a small selective part of the picture then yes it is possible for the entropy of a sub-system of the whole universe to have decreasing entropy (only at the expense of some other part that is gaining entropy at a rate that the total system agrees with the laws of thermodynamics). Life evolves and lives in the places where there is an energy gradient. Life evolves at hot under sea vents where 400 degree water (and associated high energy content molecules) meets 10 degree water (poor in high energy molecules). Life lives where a sun (6000 degrees) pours out 10^17 joules per second of energy into the cold dead maw of the night (2.7 degrees).

15. Jan 16, 2010

### edpell

p.s. we are like salmon swimming up stream

16. Jan 29, 2010

### baywax

Hi again... radiation event is just how I described the condition of a background radiation.
Please bear with me. Is ignorance a form of entropy... or potential?!
Would sound be antientropic? Sound is a result of work being done, causing a vibration radiating from the source of the work. The energy causing sound is unable to escape the planet and thus continues to do work on earth in the form of vibration that either effects matter or is sensed cognitively. It's almost as though sound generates a domino effect of work being done. Is there as much entropy involved with sound as there is with light or other conditions?

17. Jan 29, 2010

### ZapperZ

Staff Emeritus
Where did Styer and Bunn made that claim? In fact, if you look in Styer's paper, this is clearly the conclusion he came up with:

In other words, the decrease in entropy of the earth is SO SMALL, it would be difficult to detect it when compared to a dead Earth! Therefore, when we add to the entropy increase that surrounds the earth, there's a definite net increase in entropy of the universe!

One can read http://arxiv.org/abs/0903.4603" [Broken] and convince oneself that he was refining Styer's work and draw up the identical conclusion (just read the abstract, for example).

So both papers do NOT argue for a net decrease in entropy of the universe, which is what the Thermodynamics laws requires. A decrease in entropy for the Earth is not surprising or unexpected, since it is not an isolated system. But even considering that, both Styer and Bunn has shown that the decrease in entropy is almost
insignificant!

There's zero controversy here.

Zz.

Last edited by a moderator: May 4, 2017
18. Jan 29, 2010

### sylas

No.

Sound satisfies the laws of thermodynamics just like everything else.

19. Jan 29, 2010

### Gokul43201

Staff Emeritus
Zz, I think you and sylas may be misunderstanding raimo's point. Raimo is not saying that there is a controversy arising from an inherent violation of the 2nd law. Neither is he saying that one of the papers reports an overall reduction in entropy for the universe (i.e., he is not claiming that one set of authors is crackpottish). I believe he is saying that the two papers claim opposite signs for the entropy change within one particular open subsystem of the entire system.

20. Jan 29, 2010

### sylas

I realise that, and pointed out that he's wrong about that, especially in [post=2528642]msg #11[/post]. There is no open subsystem involved where the two papers give an opposite sign.

Cheers -- sylas