## Denying the Climate Catastrophe: 3. Feedbacks

This is the third chapter of an ongoing series.  Other parts of the series are here:

We ended the last chapter on the greenhouse gas theory with this:

So whence comes the catastrophe?  As mentioned in the introduction, the catastrophe comes from a second, independent theory that the Earth's climate system is dominated by strong positive feedbacks that multiply greenhouse warming many times into a catastrophe.

In this chapter, we will discuss this second, independent theory:  that the Earth's climate system is dominated by positive feedbacks.  I suppose the first question is, "What do we mean by feedback?"

In a strict sense, feedback is the connection of the output of a system to its input, creating a process that is circular:  A system creates an output based on some initial input, that output changes the system's input, which then changes its output, which then in turn changes its input, etc.

Typically, there are two types of feedback:  negative and positive.  Negative feedback is a bit like the ball in the trough in the illustration above.  If we tap the ball, it moves, but that movement creates new forces (e.g. gravity and the walls of the trough) that tend to send the ball back where it started.  Negative feedback tends to attenuate any input to a system -- meaning that for any given push on the system, the output will end up being less than one might have expected from the push.

Positive feedback is more like the ball sitting on top of the hill.   Even a small tap will send it rolling very far away, because the shape of the hill and gravity tend to push the ball even further in the direction of the tap.  Positive feedback amplifies or multiplies any input to a system, meaning that even small pushes can lead to very large results.

The climate temperature system has a mix of positive and negative feedbacks.

For example, consider cumulus clouds.  If the Earth warms, more water tends to evaporate from the oceans, and some of that water will form big fluffy white clouds.  These clouds act as an umbrella for the Earth, reflecting heat back into space.  So as more clouds form due to warming, there is a net new cooling effect that offsets some of the original warming.  The amount of warming we might have expected is smaller due to the negative feedback of cloud formation.

On the other side, consider ice and snow.  Ice and snow reflect sunlight back into space and keep the Earth cooler than it would be without the ice and snow cover.  As the world warms, ice and snow will melt and thus reflect less sunlight back into space, having the effect of warming the Earth even more.  So an initial warming leads to more warming, amplifying the effect of the initial warming.

Since we know both types of feedback exist, what we care about is the net effect -- does negative or positive feedback dominate?  In every catastrophic forecast you have seen for global warming, in nearly every climate model the IPCC uses, the authors have assumed that the climate is dominated by strong positive feedbacks that multiply incremental warming from greenhouse gasses many times.

This is the result:

As a reminder, the green line is the warming from increases in atmospheric CO2 concentration solely from the greenhouse gas effect, without any feedbacks taken into account.  It is generally agreed to be a warming rate of about 1.2C per doubling of CO2 concentrations, with which I and many (or most) science-based skeptics agree.  The other lines, then, are a variety of forecasts for warming after feedbacks are taken into account.  You can see that all these forecasts assume positive feedback, as the effect is multiplicative of the initial greenhouse gas warming (the pink, purple, and orange lines are approximately 3x, 5x, and 10x the green line, implying very high levels of positive feedback).

The pink line is the mean forecast from the 4th IPCC, implying a temperature sensitivity to CO2 of about 3C.  The purple line is the high end of the IPCC forecast band, implying a temperature sensitivity of 5C.  And the highest is not from a mathematical model per se, but from the mouth of Bill McKibben (sorry for the misspelling in the chart) who has on several occasions threatened that we could see as much as 10C of warming from CO2 by the end of the century.

Skeptics have pointed out a myriad of issues with the climate computer models that develop these forecasts, but I will leave those aside for now.  Suffice it to say that the models exclude many important aspects of the climate and are subject to hand tuning that allows modellers to produce pretty much any output they like.

But I do want to say a few words about computer models and scientific proof.  Despite what you will hear from the media, and even from the mouths of prominent alarmist scientists, computer models do not and cannot constitute "proof" of any sort.  Computer models are merely tools we use to derive the predicted values of physical parameters from complex hypotheses.  They are no different than the pen and paper computations an 18th century researcher might have made for the position of Saturn from Newton's celestial mechanics equations.  The "proof" comes when we take these predicted values and compare them against actual measurements over time and find that they are or are not accurate predictions.  Newton's laws were proved as his equations'  outputs for Saturn's position were compared to Saturn's actual measured position  (and in fact they were disproved, to a small extent, when Mercury's position did not accurately match and Einstein has to fix things a bit).  Similarly, hypotheses about global warming will be proved or disproved when the predictions of various models are compared to actual temperatures.

So we can't really get much further until we get to actual observations of the climate, which we will address in the next several chapters.  But I want to make sure that the two-part theory that leads to catastrophic global warming is clear.

This is the portion of the warming due to greenhouse gas theory:

As you can see, the portion due to greenhouse gas theory is relatively small and likely not catastrophic.  The catastrophe comes from the second independent theory that the Earth's climate system is dominated by strong  (very strong!) positive feedbacks.

It is the positive feedback that causes the catastrophe, not greenhouse gas theory.  So in debating catastrophic man-made global warming theory, we should be spending most of our time debating the theory that the climate is dominated by strong positive feedbacks, rather than debating the greenhouse gas theory.

But in fact, this does not happen in the mainstream media.  If you are an average consumer of climate news, I will be you have never heard a discussion in the media about this second theory.

And this second theory is far from settled.  If on the "settled" scale from 1-10, greenhouse gas theory is an 8 or 9, this theory of strong positive feedbacks dominating the climate is about a 2.   In fact, there is plenty of evidence that not only are scientists estimating feedbacks incorrectly, but that they don't even have the sign right and that net feedbacks may be negative.

This is a bit hard to communicate to a layman, but the positive feedbacks assumed by the most alarmist and catastrophic climate forecasts are very, very high.  Way higher than one might expect in advance upon encountering a new system.  This assumption of strong positive feedbacks is one that might even offend the sensibilities of the natural scientist.  Natural systems that are long-term stable (and certainly for all its variation the climate system has remained in a pretty narrow range for millions and millions of years) are typically not dominated by positive feedbacks, they are dominated by negative feedbacks.

If in fact our climate temperature system is dominated by negative feedbacks, the future warming forecast would actually be below the green line:

OK, without getting in and criticizing the details of these models (which would by the way be a pointless wack-a-mole game because there are dozens of them) the best way to assess the validity of these various forecasts is to now consult actual observations.  Which we will begin to do in our next chapter, part 4a on actual temperature measurements.

• kevinsdick

I'm a long-time CAGW skeptic. Even so, those charts are very eye-opening.

It actually inspired the following analogy: someone makes the catastrophic claim that, according to computer models based on well understood growth principles, antibiotic resistance is going to result in a bacterial super-slime covering the globe and wiping out civilization. Therefore, we have to stop all antibiotic use immediately.

Skeptic object that antibiotics do a lot of good, there are fundamental reasons to believe the growth models don't take into account a lot of factors, and in fact measurements show bacterial growth much slower than the models.

Catastrophe proponents scream that the skeptics don't believe in antibiotic resistance, don't believe in evolution, and don't believe in science.

• Matthew Slyfield

Even assuming that positive feedback exists, the evidence that even the IPCC's worst case for warming would have catastrophic impacts is weak at best.

• J_W_W

Most Electrical Engineers would totally understand this argument. Stable systems do not (and frankly cannot, definition of stable) have large positive feedbacks. Positive feedback exists and can be forced, but for systems that are stable, it cannot be the main driver and the net feedback for a stable system can't be dominated by positive feedback.

An the climate system, as a whole, is actually quite stable, therefore it must have fewer influential positive feedbacks and more influential negative feedbacks.

• geran

You do a great job of explaining how Earth’s systems can handle an energy imbalance. That is good science. But, you keep coming back to the AGW pseudoscience:

“It is generally agreed to be a warming rate of about 1.2C per doubling of CO2 concentrations, with which I and many (or most) science-based skeptics agree.”

There is no such “consensus” among “science-based skeptics”.

Here’s a quote from NOAA:

“Scientists say that doubling pre-industrial carbon dioxide levels will likely cause global average surface temperature to rise between 1.5° and 4.5° Celsius (2.7° to 8.1° Fahrenheit) compared to pre-industrial temperatures.”

So, your un-named source indicates: 1.2C

NOAA indicates: 1.5 to 4.5 C

Pick a number!

The point is, there is NO proof of any of these figures. “Science-based skeptics” need to see some scientific proof, or solid evidence, that CO2 "climate sensitivity" is anything other than ZERO. Otherwise, IPCC/CO2/GHE/AGW is just a “belief system”. And, a “belief-system” that fosters fraud, data-manipulation, and censorship, is a hoax.

• donald

Perhaps our little stable world is a case of a local minima. minor events have negative feedback keeping us in the happy range. But pushing a system from that local minima to the top of the local maxima... One push sends us happy back to the localized minima, but push the other way and all hell breaks loose and we all are going to learn to surf in Kansas. Or wherever the superflood reaches. of course that graph looks different than Bill's and probably is not a function of CO2, but from a math geek standpoint it's fun to think about positive and negation feedback.

• AudreyA

Just a comment on this in general. Let's say it's all true and we face global warming of several degrees or more. Not too long ago, Inuit people were thriving in the frigid Arctic while the Kung bush people lived fairly comfortable lives in the bleak deserts of Australia with nothing more than STONE-AGE tools. I think, just maybe, we can adapt.

• Zachriel

Coyote: Natural systems that are long-term stable (and certainly for all its variation the climate system has remained in a pretty narrow range for millions and millions of years)

Actually, the Earth seems to oscillate between two extremes; ice ages, and ice-free periods.

geran: “Scientists say that doubling pre-industrial carbon dioxide levels will likely cause global average surface temperature to rise between 1.5° and 4.5° Celsius (2.7° to 8.1° Fahrenheit) compared to pre-industrial temperatures.”

That includes the feedback due to increased water vapor.

• The all else being equal, 1 to 1.2°C increase per doubling of CO2 concentration is pure science, laboratory tested and repeatable and has been a known and accepted fact for a very long time. It includes no feedback, positive or negative.

• MB

> This assumption of strong positive feedbacks is one that might even offend the sensibilities of the natural scientist.
We're not talking a runaway scenario here - there's multiple stable points (attractors) that get transitioned between. That describes a wide variety of physical processes, and should offend absolutely nobody.

For anyone who wants to actually understand some basics, I'd suggest this quick article on Ars (which, though focuses on climate science, has absolutely nothing specific about it)[1]. Further info is just a Google or Wikipedia away.

> The models exclude many important aspects of the climate
Do they exclude things? Yes - of course, they're models. Do they exclude important things? None that we know of - and as we add more and more things, the general answer doesn't change.

> [The models] are subject to hand tuning that allows modellers to produce pretty much any output they like
You give a lot of credit to modellers being able to predict the outcome of their model based on parameterizations (if they could do that, they wouldn't need the model).

Some more background - yes, some things are "tuned" with parameters - these are things that are infeasible for the model to calculate in a reasonable timeframe because of size or complexity. We don't believe that it's important to the accuracy of the model that we calculate how much solar radiation hits a particular grain of sand in the Arizona desert; though we do believe that some overall solar radiation parameters are important.

Each parameters are only plausible in certain ranges (the plausible ranges are based on other science), which gives us some bounded conditions. Even so, considering the chaotic behavior and number of parameters - it is feasible that using the wrong value could give wildly different results.

That's one of the purposes of the ensemble approach of running multiple iterations and comparing outputs. Ensembles are used for different models, different parameter values, different initial conditions, etc. - current ensembles agree that, regardless of initial conditions or parameter values, there is a net positive feedback. Those ensemble results are basically what is reported by the IPCC.

If you wish to know more about parameterization of models, (and even help narrow down some uncertainties!) see climateprediction.net[2].

> Despite what you will hear from the media, and even from the mouths of prominent alarmist scientists, computer models do not and cannot constitute "proof" of any sort.

You're conflating two different aspects of modeling. On the one hand, we must provide proof of the model - this is done by matching observations to the model, and has been done time and time again. Once the model is "proven", we can then use that model to project the future. This doesn't really constitute a "proof" of the future (which, philosophically, I think would be impossible) but is the basis of all science - we use the model of gravity to project how to put a satellite in orbit.

In some cases, we actually do accept models as proof as well, if they are consistent with a large variety of other observations. Eg., black holes cannot be directly observed - but we accept them, because the model says they're there and we find other observations consistent with that. Similarly, we can't actually see electrons and such - we deduce that they're there because the fit the model, and other observations we can make are consistent with them being there[3].

This starts to get annoyingly philosophical at this point - so I'll leave it at, yes models can provide proof (note that nothing can constitute proof on its own).

> Similarly, hypotheses about global warming will be proved or disproved when the predictions of various models are compared to actual temperatures.

No - we can use hindcasting to provide proof of models without waiting until the future (though, of course, we also compare things in the future). The idea that we can only show evidence for the model working by waiting until the future is ludicrous - we don't have to wait until tomorrow to accept that the model of planetary bodies says the sun will rise.

> For all its variation the climate system has remained in a pretty narrow range for millions and millions of years.
Define "pretty narrow range" and relate to temperatures humans have successfully adapted to over the last 10,000 years that we've been here. Also, extrapolate the extremes of your "pretty narrow range" into potential effects of current human civilization.

If you don't think the Earth's "pretty narrow ranges" over the last "millions and millions of years" have the potential to royally screw things up - then understanding climate models is the least of your problems. As a start, consider how much of Florida has been underwater at some point in the last "millions and millions of years"[4] and the 20 million people that live there in the now.

Ice age cycles of the past 800,000 years have seen oscillations of temperature of 5-6C peak to trough. The forcing for these oscillations is provided by a very weak change in the amount of summer sunlight in the northern hemisphere, which by itself would provide a negligible change in energy flux. The changes in temperature are the result of ice and snow cover changes which alter the absorption of sunlight by the earth's surface and greenhouse gas emission and absorption by the oceans in response to the temperature change, and changes in dust particles in the atmosphere. All of these phenomena amount to positive feedbacks which have caused enormous changes in the climate of the earth. This refutes the claim that climate feedbacks on the earth are intriniscally small.

The climate system has been oscillating between ice ages and interglacial periods with a peak to trough difference of about 5 to 6C. The trigger for the oscillations was a small change in summer light intensity in the northern hemisphere due to variations in the earths elliptical orbit and axial orentiation. The change in net energy flux due to these insolation changes was small. The changes in energy 'flux were amplified by changes in snow and ice cover, and GHG concentration in the atmosphere. This caused the temperature changes of 5 to 6C. This is evidence of large positive feedbacks in the earth atmosphere system.

Feedbacks in the climate system caused nearly all of the temperature changes during the ice age cycles which resulted in huge.changes in sea level.

• Matthew Slyfield

Which occurred over many thousands of years.

Sea level change has been steady at a few mm per year for the last couple of centuries and shows no sign of acceleration.

Only land bound ice can cause sea level change. Yes, there is enough ice in the Antarctic ice cap for a couple hundred meters of sea level rise. However, if you imagine that there is any scenario under which it could melt in less than several thousand years, you are very wrong.

"Only land bound ice can cause sea level change.
That statement is wrong. The fact that water expands when heated except when it is near the freezing point is taught in middle school. Measurements show that the oceans have been gaining energy. Check out the -2000M plot of ocean heat content on this NASA web site.
https://www.nodc.noaa.gov/OC5/3M_HEAT_CONTENT/
Here is a plot of sea level which shows acceleration of sea level rise in recent times.
https://robertscribbler.files.wordpress.com/2015/05/hansen-sea-level-rise.png
In fact sea level is an excellent thermometer for the earth, because most of the heat gained by the earth gets stored in the ocean.

• Matthew Slyfield

1. There are finite limits to thermal expansion of the oceans.

2. The bulk of the oceans volume, is barely above freezing.

3. 3.1mm per year is insignificant. The alarmists are predicting 10s of meters of sea level rise by the end of the century. even at 3.1 mm per year, That's still over 300 years for just 1meter of sea level rise.

4. Sea level varies by meters over both space and time, on short time spans (tides and waves). The idea that sea level can be measure to a precision of mms is absurd.

1. This is an empty and meaningless statement. There is a large volume of water in the oceans to expand. The volume doesn't have to reach infinity to flood low lying areas of seashore.
2. Not so. Salinity and pressure lower the freezing point of water by a lot. At standard temperature and pressure, a saturated salt solution freezes a about -20C instead of 0c. The only place the ocean freezes is at the surface in the polar regions. The vast tropical oceans are nowhere near freezing. Check out the typical tropical ocean temperature curve.
https://en.wikipedia.org/wiki/Thermocline
3. You are assuming that the increase will remain at 3.1mm/year. In fact it will continue to accelerate as it has been doing. Even a 30cm increase, which would occur by 2100 even without acceleration would damage some low lying populated areas and create millions of climate refugees.
4. Average sea level changes can be measured. Scientists who study this don't consider it absurd. What is absurd is your notions about science.

• Matthew Slyfield

3. That graph doesn't show continuing acceleration. It shows two point changes in the rise in rate. 30 cm over a century is easily dealt with by minor inland migration, it's not a catastrophe.

You previously said there was no acceleration. Now you are debating the form of the acceleration. The straight lines are estimated to connect what appear to be change points. There is no assurance based on these lines that acceleration will not continue, which is what you appear to be assuming.

The projected sea level rise will continue accelerate according to all long term projections, just as it has accelerated recently. Here is one recent paper which gives the details of the sources of past sea level rise based on detailed observations and projects the range of future sea level rise.
It includes thermal expansion ( which you incorrectly claimed was not a factor), mountain glaciers, Greenland and Antarctic glaciers and human storage of water.
http://www.pnas.org/content/113/10/2597.full
It shows that the median sea level rise by 2100 depends strongly on emissions scenario. The mean projection of sea level rise for the lowest emissions scenario is about .3M and about 0.8M for the high emissions scenarion. The sea level curves show no signs of saturation by the year 2100 and an accelerated trend will continue beyond 2100. There are similar publications which give similar estimates and the IPCC AR5 report developed a consensus estimate from those papers. Some researchers give higher sea level rise projections.

• Edward Bo

Wow! The recent inflection point in that graph just happens to occur right when they decided to report the data from a different measurement system. What a coincidence!

In my professional technical work, I'd get fired for that bait and switch reporting. The fact is that if you use the same tidal gauges that were used throughout the 20th Century, there is NO acceleration. A recent report from NOAA concludes that recent sea level rise is under 1.5 mm/year, essentially the same as it's been for a century.

As to the early 20th Century warming (~1910 - 1945), according to the mainstream climate science analysis, it cannot be explained by the small increases in CO2 that were occurring then.

And then the 1945 - 1975 cooling (somehow adjusted recently to a "pause") that you can see clearly in the slowdown in sea level rise... Oh, wait, you can't see it??!! Maybe the relationship between temperatures and sea levels is not so simple.

But of course, we must believe the predictions that all of a sudden sea level rise will hit 10 mm/year or more!

• Edward Bo

Really? You can't figure out that there's a huge difference in snow/ice albedo feedback when you have mile-high glaciers down to 45N latitude as during glacial periods, and sea ice at 75N as during interglacial periods like the present?

Take a good look at the Greenland ice core data, and the huge temperature swings of the Dansgaard-Oeschger events during glacial periods compared to the much smaller swings of the comparable Bond cycles during interglacial periods.

Ask yourself why, in cycle after cycle, the rapid warming out of glacial periods stops suddenly at temperature levels very similar to those of the present. If positive feedbacks were still dominant, the warming would have continued.

The first half of the 20th century saw warming due to increased solar irradience. CO2 rise was not sufficient to explain that. Everyone who looks into this question agrees on that.

" A recent report from NOAA concludes that recent sea level rise is under 1.5 mm/year, essentially the same as it's been for a century."????
This NOAA report done 7 months ago says the opposite:.
https://www.climate.gov/news-features/understanding-climate/climate-change-global-sea-level
"The pace of global sea level rise almost doubled from 1.7 mm/year throughout most of the twentieth century to 3.2 mm/year since 1993."

In fact the recent rate of rise has been checked by independent means in additon to tide guages, as described in a figure caption for the last graph shown in the article:

"Independent estimates of the relative contribution of thermal expansion (orange line) and melting and other added water (blue line) to global sea level rise add up (purple line) very closely to the observed global sea level rise measured by tide guages (black line). Graph adapted from Figure 3.27 in the BAMS State of the Climate in 2014 report."

To be sure, the albedo feedbacks in this period are smaller than during ice age cycles. Water vapor seems to be an important factor which doubles the effect due to long lived GHG's emitted into the atmosphere by human activity.

D-O events were over quite quickly. It took about 800 years before reduced albedo warmed the earth for long enough periods for the ocean to emit enough GHG's to provide additonal warming feedback. The D-O events did not last long enough to do that, and whatever it was, the stimulus for the changes in glaciation disappeared. The change in insolation due to the alignment of the earths orbital perihelion and axial tilt with the NH summer lasts long enough to keep the interglacial periods going for 10's of thousands of years.

• Alan_McIntire

There were no ice ages for millions of years, until the Pleistocene, probably due to continental arrangement.
The sun started out only about 70% as luminous as it is now Despite that, there have been liquid oceans, and life, on earth for just about 4 billion years an indication of large NEGaT/iVE feedback.
There's also the Stefan-Boltzmann constant, showing that luminosity incrases as the 4th powere of temperature. That means that as luminosity increases 16 times, absolute temperature only doubles.
Anyone believing in those large positive feedbacks is an astrophysics andk mathematics denier.

• dbw

You discuss a preference to should global temperature vs. CO2 in your example to illustrate how other variables affect temperature. This is a conceptual example not based on actual data. It is useful to compare data for global temperature (Y axis) vs. CO2, using e.g., NASA GISS and Mauna Loa (ML). See posts in WUWT here https://wattsupwiththat.com/2014/09/12/a-look-at-carbon-dioxide-vs-global-temperature/ and here https://wattsupwiththat.com/2015/04/22/a-statistical-definition-of-the-hiatus-in-global-warming-using-nasa-giss-and-mlo-data/. The alarmists will argue that this is invalid since we know there are other variables. However, we also know that the storyline models do not account for other variables - e.g., moisture, aerosols, solar forcing, .. saying that feedbacks are caused by CO2 so it doesn't matter if they are properly accounted for. Well... it sure does matter if you account for other variables since the feedback effects are much larger than the direct effects.

We know that the integrated assessment climate models (IAMs) are deterministic physical models of the climate with predetermined built in physical cause and effect structures. From the data since the late 1990s and during the period 1940s through 1970s we can say they are wrong based on their ability to explain the data (facts). The lack of correlation in temperature vs. CO2 is despite the vehemently said claim that manmade CO2 is the primary cause of global warming. In fact plots such as the ones referenced demonstrate that since the late 1990s the complete lack of correlation allows us to conclude that the assumption that CO2 is the major thing driving global mean temperature is not just a lousy hypothesis, it’s flat out wrong and unsupported by the data (facts). We can say that during this period all of the variability (scatter) in the data is due to “NOT CO2.”

"The sun started out only about 70% as luminous as it is now Despite that, there have been liquid oceans, and life, on earth for just about 4 billion years an indication of large NEGaT/iVE feedback."
This ancient era is a big mystery to paleoclimate scientists. It doesn't make sense to conclude anything about feedbacks in the modern climate system based on it. Many different theories have been proposed to explain why liquid water was there and none have been strong enough to take hold.

http://hea-www.cfa.harvard.edu/lifeandthecosmos/wkshop/sep2012/present/faint-young-sun-smithsonian-sep12.pdf

"From the data since the late 1990s and during the period 1940s through 1970s we can say they are wrong based on their ability to explain the data (facts)."
The models have forcings as their inputs and need to account for the effect of clouds in an empirical fashion. One of the problems with forcings is the estimation of the effects of human caused aerosals. In addition the models don't do a good job on ENSO. Despite this, they are able to reproduct the outline of 20th century climate change. What they do get correct is the effect of CO2 and water vapor feedback in clear air.
Here is an example of a recent paper from China
http://www.sciencedirect.com/science/article/pii/S1674927813500050
which shows good agreement between multiple model runs and the observed global average temperature.
http://ars.els-cdn.com/content/image/1-s2.0-S1674927813500050-gr2.jpg

• Ben Winchester

An the climate system, as a whole, is actually quite stable, therefore it must have fewer influential positive feedbacks and more influential negative feedbacks.

J_W_W, yes.

There's an incredibly important and very large negative feedback in how bodies radiate away energy: the outgoing power scales with T^4. That's the Stefan-Boltzmann Law. This negative feedback dominates; it's why our planet and climate exists at all.

When climate scientists talk about positive feedbacks, they're talking about the rest of the feedbacks; all the net feedbacks except the Stefan-Boltzmann Law. Warren Meyer seems to be ignorant of this, which is... uhh, let's just say "weird". It's a pretty important, pretty basic point behind the physics of climate science.

When that's accounted for, there's nothing odd about the feedbacks climate scientists find.

The biggest one is water vapor. We know that air can hold 7% more water per degree C. From research, climate scientists expect that the actual absolute humidity will increase by about 4% with each additional degree, or equivalently that relative humidity will stay about the same. The "feedback" here is really what you get if you just hold the humidity constant.

• Edward Bo

Here's the NOAA report that concludes recent sea level rise is 1 - 2 mm/year, confirmed by multiple methods:

http://www.star.nesdis.noaa.gov/sod/lsa/SeaLevelRise/documents/NOAA_NESDIS_Sea_Level_Rise_Budget_Report_2012rev1.pdf

Obviously too much of a challenge to the alarmists for them to continue to stand behind it...

Yes, I'm cynical. When you see the IPCC Summary by (I mean for) Policymakers blindly compare the earlier lower tide gauge trend to the newer higher satellite trend, then bury hundreds of pages into the main document the fact that the tide gauge trend shows no acceleration, you realize it's a political document and not a scientific one.

By the way, do you have any source for your claim that there was a significant increase in solar insolation early in the 20th Century? All of the sources I see show no more than 0.1 W/m2 variation, or say not enough resolution in the measurements to say anything definitive.

• Edward Bo

The seasonal change in insolation due to orbital precession is far greater than the effect of recently increased CO2. The last time I checked the GISS-E model source code, I was astounded to see that they took insolation as constant over the year, assuming a circular, not elliptical orbit, and negating this effect. (I don't know if more recent versions have corrected this.)

You miss my point about D-O events. As short as they were, they produced far more temperature change than Bond cycles of the same periodicity in interglacial periods. This is very strong evidence that sensitivity in interglacial periods such as the present is far lower than in glacial periods.

• Ben Winchester

It would still implicitly include the feedback from radiation; the Stefan-Boltzmann Law.

That's what the author left out in the discussion. There is always that one large negative feedback that dominates overall; without it the Earth would be millions of degrees. And when climate scientists say "positive feedbacks", they're talking about the sum of all the other feedbacks.

And you have to denial pretty hard to dismiss water vapor as a feedback. We've known for almost 2 centuries that warmer air holds more water vapor. It's very basic physics, and very-well verified.

It seems that the author of the report you quoted, Eric Leuliette, changed his mind about the recent rise in sea level in a report he produced in the year following the one you quoted. The new sea level rise he calculates for 2005-2013 is 3.0+/- 0.4mm/year.
http://www.star.nesdis.noaa.gov/sod/lsa/SeaLevelRise/documents/NOAA_NESDIS_Sea_Level_Rise_Budget_Report_2014.pdf
I guess this is the source of the confusion in the AR5 sea level numbers.
I haven't had the time to read both papers carefully enough to see which estimates changed in the year between the two reports.

I didn't miss your point. I agreed that the albedo feedback potential was much larger during the periods of glaciation.

• Black body radiation - ultimately that's the sum total of all energy transfer on a universal basis. But it's not a feedback. Feedback is a response to a given input, a system wide change. Radiation is a mechanism.

People understand on a visceral level heat transfer via conduction and convection. Radiative transfer, on the other hand, requires a more intellectual understanding and the whole greenhouse gas mechanism is based on radiative transfer. I think that's part of what makes it so hard for most people to understand what's actually going on.

Yes, warmer air holds more water vapor and water vapor is a much stronger greenhouse gas than CO2. But it also causes more clouds and clouds increase the albedo which is a negative feedback. So even that isn't as simple as it sounds.

• The only thing hindcasting can do it to prove your formulas do a good job of curve fitting. Given enough terms, you can fit a formula to any dataset. It doesn't *prove* anything about their usefulness for predicting the future. That can only be done by comparing the predictions to reality when that time comes around. If the predictions match the observations, and continue to do so, the strength of the underlying theory is enhanced. If the observations don't match the predictions, then the model is wrong.

• Radiation is a mechanism, not a feedback. Feedback is a systemic response to an input.

Increased water vapor also increases clouds which is a negative feedback - it increases the albedo which reflects the radiation from the sun back out. So even water feedback isn't that simple. Plus there are observational datasets that are showing a *reduced* relative humidity, not constant.

• Given enough terms, you can fit a computer model to any dataset. It's accuracy can only be assessed when it is used for predictions and then measured against *that* data. If the predictions match the observed data at that time, then the perceived accuracy underlying model is increased. If the predictions do not match the observations, then the model is wrong, no matter how well it fits historical data.

Models, in and of themselves, *cannot* prove anything.

• The volumetric difference in water from thermal expansion at Earth sea normal surface temperature from a 1°C increase in temperature is about 0.04% and that changes so little as to be insignificant. For all intents and purposes over the range in temperature we're talking about, it's a constant. Going from 4°C (maximum density) to 35°C (far higher than the ocean is going to get under any scenario) is a volumetric increase of about 1%

While the average surface temperature (what we normally think of as its temperature) of the ocean is about 17°C, the vast bulk of it is between 0°C and 3°C which is actually *below* maximum density. Warming that water by even a degree or two will actually *decrease* it's volume. So the effect of melting sea ice has no real effect on sea levels.

The melting of land ice is what raises sea level. Well, that and land subsidence and erosion has the same effect for all human purposes. But the oceans have been rising throughout human history and we've dealt with it just fine, adapting and moving as needed. Sure, things change. They always change. That's why we have been so successful as a species. We adapt to changing circumstances.

Over the course of 100 years, the cumulative difference between a 1mm rate and a 4mm rate would be less than 12 inches. Simple math.

• Ben Winchester

Feedback is a response to a given input, a system wide change. Radiation is a mechanism.

Yes. Radiation is the mechanism by which the dominant source of feedback to temperature occurs in the Earth's climate system.

I mean, if you want to be technical, you could say that the radiation itself isn't the feedback; the feedback is the increase in radiation that results from an increase in temperature. In other words, the feedback is defined by the derivative of the radiation with respect to temperature. But I figured if I just said "the feedback from the radiation", you'd know what I meant. Not the radiation itself; but the way the radiation changes with temperature.

It's a very large negative feedback, and is why scientists don't actually find the net feedbacks of climate to be anything whacky. And it's, ummmm, "odd" that Warren Meyer left this out of his discussion. It's kinda a key point.

Yes, warmer air holds more water vapor and water vapor is a much stronger greenhouse gas than CO2. But it also causes more clouds and clouds increase the albedo which is a negative feedback.

How does that work? Clouds aren't water vapor, clouds are liquid water. They're made up of water that condensed out of its gaseous form. Condensation occurs only after the air has cooled down below the dew point, right?

But because warmer air can hold more water vapor (i.e., in gaseous form), a fixed amount of water in warmer air would "prefer" to stay in that gaseous form. It'll a harder time condensing out and forming clouds. This is what happens if you don't let the total water vapor increase as air warms: fewer clouds.

However... if you keep the relative humidity the same, then clouds will form about as easily as before. (Relative humidity is how saturated the air is, as measured relative to the maximum saturation. If you're holding relative humidity fixed, then the air will hold about 7% more actual water for each degree warmer).

In terms of a fixed relative humidity, reaching 100% humidity always requires the same amount of temperature drop. Air at 50% relative humidity takes about a 10C temperature drop to reach 100% humidity, no matter what temperature is started out at. And clouds form when you reach >= 100% humidity.

Since temperature changes are the dominant factor for determining when clouds form... this means that clouds will form about as easily as before, even with a fixed relative humidity, even with increasing total water vapor in the air.

TL;DR: warmer air holds more water vapor, and because of that, the liquid-water clouds will form about as often with warming, so long as relative humidity doesn't actually increase. Climate scientists find that relative humidity will probably stay about constant.

• DavidAppell

Feedbacks aren't "assumed" -- they are a consequence of the physics.

And the biggest ones are pretty obvious.

• Please remember that I'm writing, not just for you, but for anyone that reads this so I'm probably repeating a lot of stuff you already know well.

Greenhouse gases (GHGs) absorb and then re-radiate in a random direction (meaning half up and half down) radiation in the longwave IR spectrum. Half of whatever is heading outwards at the pertinent wavelengths that hits a greenhouse gas molecule at goes back down.

CO2 Absorption centers and blackbody radiation -
CO2 Absorption Wavelengths in nm: 15,000, 10,400, 9,400, 4,300
Solar output peaks at about 800nm and is pretty well ended by about 3,000nm

Now - there is a whole lot of empty space between molecules of all the gases. Very few of the molecules that are there absorb in the IR spectrum. So we're talking a fraction of a fraction of a fraction of the total radiation that the GHGs affect.

In and of itself, CO2 actually has only a fairly minor effect on net temperature. For each doubling of concentration, the overall effect is an increase in ambient temperature of about 1.1°C. Over the last 150 years or so it has gone from 280 ppm to 400 ppm. That accounts for an increase of about 0.47°C due to CO2. Everything else about the catastrophic anthropomorphic (human caused) global warming theory is supposed to be a secondary positive feedback effect.

Water vapor is responsible for over 90% of the atmosphere's greenhouse effect. Higher temperatures means more water evaporates. Not a lot of difference from 0.47°C, but it's there. But water is actually a very complex molecule in all its various forms and interactions with the climate.

* Unlike CO2, the amount that is in the air at any place can change dramatically from day to day all over the planet.
* Unlike CO2, it can condense into a liquid releasing latent heat which can later evaporate into vapor again absorbing latent heat.
* Unlike CO2 it is lighter than air so it convects upward.
* Unlike CO2 it can reflect sunlight when it condenses at altitude or falls as snow on the ground.
* Unlike CO2, the IPCC admits that water vapor, the predominant greenhouse gas providing over 90% of the greenhouse effect, is TOO COMPLEX to actually model so they instead just apply some linear coefficients to represent it .... in a model for a non-linear chaotic system.

If water has a net positive feedback loop on temperature, the Earth would have become uninhabitable long ago - long before there was anyone or much of anything to try and inhabit it. So it's got to be a net negative.

There are other ways human activities impact climate, at least on a local basis. A couple of points mentioned in other installments of this analysis include particulate fallout from coal and wood burning darkens snow and ice leading to reduced albedo (less reflecting incoming light back out) and deforestation changes the albedo, wind patterns and CO2 absorption by plant life (Kilimanjaro's glacier is a great example. It was shrinking when they clear cut or slash and burned the areas around it. Then they started to let the forests grow back. Now it's growing again).

So in short:
* Has the climate warmed in the last 150 years? Yes
* Has the climate warmed in the last 50 years? Yes
* Is CO2 a Greenhouse Gas? Yes
* Have we put more CO2 into the atmosphere than there would have been otherwise? Yes
* Have our actions caused at least some warming? Yes
* Are we the sole cause of all the warming in the last 50 years? No
* Are we even the majority cause of all the warming in the last 50 years? Very doubtful

The fact is that climate naturally changes. It always has and it always will. There's nothing we can do about that. We've just scratched the surface of being able to understand and measure all of the forces and mechanisms that come into play. We're discovering new things about the Earth all the time. We don't even know all the right questions to ask yet.

It is a chaotic system where something totally inconsequential in one place makes for huge changes in another, yet other seemingly huge changes don't seem to make any difference at all. Our time perspective is so miniscule as to be laughable. We envision ourselves to be far more powerful than we are and our egos make us imagine we can control things that are still far beyond our means.

That doesn't mean we should just give up and go live in a cave. We are learning and will continue to do so. Our abilities will continue to increase. Hopefully our wisdom on when and how to use them will also continue to increase. But we need to stay grounded in reality and honestly appraise what we have and know now. Be aware of our weaknesses as well as our strengths and work with what we have.

• Edward Bo

Diane:

Sea water (~3.5% salt) does not have a negative coefficient of expansion below 4C, as fresh water does. The tables do show the coefficient is virtually zero below 5C.

The values given in any table I've seen, for both fresh water and sea water, are given for an ambient pressure of 1 atm. I haven't found any good source for how these vary with ambient pressure.

• Edward Bo

For decades, most scientists who looked at the possible warming effect of increased CO2 believed it would not be significant, because the warming would bring increased water vapor (slight additional warming) which would bring additional clouds (significant cooling.

Basic physics, after all!

• Edward Bo

Yes, the financial world is littered with market models that hvae been tunded to hindcast very well, but turn out to have absolutely no predictive value.

In order to claim you know the rate of increase in sea level height, you will have to do a more detailed calculation than what you have posted. Most of the ocean is tropical and above 4C. The numbers you have quoted are applicable to normal pressures and pure water. Salt water is different. Your simple math is insufficient.

Are you claiming the scientists who have spent their working lives studying this have failed to do the "simple math"? This seems like the Dunning Kruger effect is working in your brain. They are saying that thermal expansion is about 1/2 of sea level rise.

An average increase of 12 inches doesn't tell the whole story. Sea level rise will not be equal everywhere, and it could effect low lying areas in Florida and Bangladesh very severly. In addition sea level rise rates are accelerating and sea level rise is expected to continue after 2100.

• Ben Winchester

Please remember that I'm writing, not just for you, but for anyone that reads this so I'm probably repeating a lot of stuff you already know well.

Sure! And likewise. Plus, I have no idea what your background is, and I suppose you don't know much about mine, either.

Diane: Now - there is a whole lot of empty space between molecules of all the gases. Very few of the molecules that are there absorb in the IR spectrum. So we're talking a fraction of a fraction of a fraction of the total radiation that the GHGs affect

That's not quite right. The Earth's surface receives about 2x as much energy from back-radiation as it does directly from the Sun. To put it another way, only 1/3 of the surface's energy comes directly from the sun, versus coming from backradiation by greenhouse gases. The greenhouse gases play an absolutely crucial role there: without them, the Earth would be uninhabitable. It'd be an ice planet.

There really is a *lot* of atmosphere above us, and a fair chunk of it absorbs the outgoing radiation pretty well. And any outgoing energy that's absorbed is 1000x more likely to be passed around to one of those non-emitting non-greenhouse gases than it is to be re-emitted. So it takes some time for heat to successfully leave our greenhouse-gas-covered earth. I'd be happy to get more into the details of this if you're interested; it'd be a learning experience for me, too. We can look up the measurements for optical depth for the different greenhouse gases and see how the scientists get their results. There's some good software out there to that purpose (e.g., MODTRAN).

Diane: Water vapor is responsible for over 90% of the atmosphere's greenhouse effect.

I usually hear more like 60-70% for water vapor. CO2 is about 15-25%, and then you also have methane, ozone, etc. But the key thing about water vapor is that while, yes, it's an important greenhouse gas, there would be very little of it in the air without the other greenhouse gases. As you point out, it's the only one of the greenhouse gases that condenses out at Earth temperatures -- and in fact, it's the only major greenhouse gas with a major dependence on temperature. That means that it's driven by the other greenhouse gases, for the exact reason you pointed out.

Diane* Unlike CO2, the IPCC admits that water vapor, the predominant greenhouse gas providing over 90% of the greenhouse effect, is TOO COMPLEX to actually model so they instead just apply some linear coefficients to represent it .... in a model for a non-linear chaotic system.

Hmm. Citation needed. Or really... I'm sorry, but this is just not true.

In some places linear coefficients would be fine: e.g., in linear formulas. But in others they aren't, and no, they aren't used there. For instance, the dependence of temperature on evaporation. So if you can back up this claim with a citation, I'd appreciate it.

Diane: If water has a net positive feedback loop on temperature, the Earth would have become uninhabitable long ago - long before there was anyone or much of anything to try and inhabit it. So it's got to be a net negative.

No; the water itself can have a net positive feedback loop, so long as there are other negative feedbacks that limit it. Which there are; the Stefan-Boltzmann Law is the dominant one. It's what makes the climate generally stable instead of chaotic.

Diane: The fact is that climate naturally changes. It always has and it always will.

[Citation needed] again, sorry. Most reconstructions of past climates show that the climate is actually usually pretty stable. It changes relatively quickly from time to time, like when the Earth is hit by a meteor or we're going in or out of a glacial period. But it's not always changing. The present change is the fastest in the last few thousand years. Over the Holocene, the period in which civilization started and mankind flourished, the climate has actually been quite stable indeed, with only gradual changes. (Look for Marcott et al 2013, which is the most comprehensive paleoclimate reconstruction to date).

Note that if the climate *was* chaotic, with large and fast fluctuations, mankind would have had an immensely difficult time forming a civilization. Growing zones would change markedly from year to year. Much of the Earth would be alternating between crazy weather extremes, rainforest-like deluges one decade followed by desert conditions for the next, and then back again. Or with wild swings of hot and cold, like Miami vs Toronto.
We wouldn't be able to grow anything. That's what "chaotic climate" would look like. And of course we do see some weather extremes, but they're usually regional, smaller, and they aren't the norm.

By the way, please, if there are any questions you have or any of my claims you doubt, please ask your questions and ask for citations! I'm quite used to being asked to back up my claims (it's the MO in grad school), so I don't take it personally. It's all part of fair, critical examination. Hell, I'd be almost offended if you didn't ask for citations: it might imply that you were just going to brush me off instead of giving my arguments a serious look with a critical eye.

If you're interested, I can recommend an excellent introductory textbook on the subject of climate: Ray Pierrehumbert's "Principles of Planetary Climate". It discusses a lot of what you brought up here, about how radiation and greenhouse gases work, and cites the experimental data and math to back it up. This stuff isn't new; scientists were measuring the optical properties of gases back in the 1800s. (Also, you can find this book online for free; google it).

• Red Black

Yet the majority of climate models have run hot when compared to observation over the past two decades. This would suggest the net feedback is off and not correctly modeled.

• I'm not saying that sea level rise has nothing to do with temperature. It does, unquestionably. But the average rate of increase has gone from 1.8 mm/yr in the 1800's to as much as 3.1 mm/yr in some of the more recent measurements and places. That's an increase in the rate of a little over 1.3 mm/yr of sea level rise. The sea level would be rising even if we had absolutely no effect on the climate.

There are 25.4 mm/inch. So for each increase of 1mm/yr there is a cumulative additional 4 inches of sea level rise over the course of 100 years. If the rate of increase immediately doubled from what it is today, that would still only mean an additional 12 inches of sea level rise.

New York City says it has experienced 12 inches of cumulative sea level increase since 1900 (higher than average). And most of that increase had nothing to do with the increase in temperatures that are being attributed to man made catastrophic global warming by warmists.

Yet Manhattan actually has MORE land area than it did in 1900. How? We built it up over time. We adapted. And that's only one potential mode of adaptation.

Did it cost money to do? Sure. But most New Yorkers never gave it a second thought. No one had to leave.

What will get done where? Depends on local conditions and what the land is worth. We're not going to wake up tomorrow morning with a flooded world. To the extent that any temperature increase is actually tied whatever we do or have done, the increase still just isn't catastrophic.

• True. The dissolved salts do change the coefficient of expansion.

For pressure effects, here a graph I found: http://www.engineeringtoolbox.com/fluid-density-temperature-pressure-d_309.html It covers both temperature and density with respect to pressure curves. It's for pure water but the percentage difference is pretty minor (about 2.7%). Good for ballpark estimates.

Pressure increases about 10 bars for every 100 meters (about 328 ft). Volume, of course, goes up when density goes down.

"The sea level would be rising even if we had absolutely no effect on the climate."
Sea level rise, strictly speaking can occur with little or no surface temperature rise, if there is an energy flux imbalance, and the energy is mixed into the deeper parts of the ocean as in a La Nina. The best estimates that we have are that there is on average, an energy imbalance of .4-.6W/M^2.
There are other sources of local sea level rise, including ongoing isostatic changes in the wake of glacial recession.

"To the extent that any temperature increase is actually tied whatever we do or have done, the increase still just isn't catastrophic.
It will probably be catastrophic for some low lying populated areas, including Florida and Bangladesh. There is a high degree of uncertainty in this, with different scientific groups predicting different outcomes. The IPCC projections are conservative. Some groups project that glacial melting will accelerate and produce sea level rises of about 2 meters by 2100.
https://www.washingtonpost.com/news/capital-weather-gang/wp/2016/03/30/what-6-feet-of-sea-level-rise-looks-like-for-our-vulnerable-coastal-cities/