bdgwx

I was going through my Parametrization Schemes book by David Stensrud to try and better understand how GCMs handle water vapor. It's pretty complicated. Quite frankly...over my head. But I do see that many of the physics modules use the Clausius-Clapeyron and other relationships. I think if there was anything wrong with our understanding of the thermodynamic nature of water vapor and associated feedbacks then it would have had an impairing effect on weather forecasts and would have been noticed long ago.

It is an amplifying feedback only. Since it does not force the climate I believe its amplifying effect is included in the C per W/m^2 part. The best analogy here are audio amplifiers. There are many agents that can catalyze a sound: drummer, guitarist, vocalist, etc. Each agent has their effect amplified by the same fixed amount, but the amplifier does not catalyze a sound on its own. The final noise level is a product of all individually amplified agents. Like with climate forcing agents it is convenient to quantify each noise forcing agent's contribution to the final sound output with the amplification factor already included.

How do you know General Relativity is correct? It is but a model of gravity. I believe that model is correct because it makes predictions of the precession of Mercury, time dilation of the GPS satellites, etc. that match observations within a reasonable margin of error. It is the same with the consensus theory of climate change. We believe the models are correct because they makes predictions that match observations within a reasonable margin of error. The natural-only or natural-mostly hypothesis makes predictions that deviate from observations by an unacceptable amount. The predictions are bad enough that they are off by an order of magnitude in predicting the EEI at least after 1950. In fact, they are so bad that they often cannot even predict the sign of the temperature change. Ya know...I'm skeptical of quantum mechanics and general relativity. Between the two they make what is often called the worst prediction in all of science regarding the cosmological constant. One or both of them is wrong by an amount so astonishing that we cannot even fathom it. But I still think QM and GR are useful and they are certainly better than nothing. The same can be said for the consensus theory of climate change. The models deployed are not perfect. They never will be. But they are undeniably useful and represent the best of what we have. It is okay to advocate for natural-only or natural-mostly models for the post WWII era. But to convince cranky skeptics like me to use them in favor of what is already available you have to demonstrate that 1) they are testable, 2) they make useful predictions, and 3) that they match reality better and/or are simpler in answering certain questions than what we already have.

I'm sure there has been a positive energy imbalance since 1800. The question is...what contributing factors caused the imbalance and how and when did their contributions ebb and flow? Some of the warming is natural especially prior to 1950. But after 1950 the net of all naturally modulated factors is far too low to explain the warming. It is so low, in fact, that naturally modulated factors may even be working to lower the EEI and thus put a cooling pressure on the climate. But when scientists consider anthroprogenically modulated factors along with the naturally modulated factors we get a reasonable match between expectation and observation. Anthroprogenic factors dwarf natural factors by about an order of magnitude...at least after 1950 when the warming became most acute. We've already had about 1.2C of warming since about 1850 and that's with about 1.5xCO2 and a +0.87 W/m^2 EEI that still hasn't equilibrated yet. Even if CO2 concentrations stabilized at the current 410 ppm level we still have several tenths of degree C to warm to work off that EEI and measure the ECS. So the question for you is this...if CO2 is only 1/3'ish effective as the consensus then where are you going to get the other 2/3 energy required to produce the amount of warming we observe?

Even Arrhenius understood the water vapor feedback and considered it in his primitive 1896 model. The ERA5 850mb data is intriguing indeed. I'd like to learn more about this decline. I'm not as ready and willing to chalk up things I don't understand to the inadequacies of the data. ERA5 is considered to be the best of the best. I'll see if I can do some digging on that 850 mb specific humidity decline from 1980-2000.

I'm just being pedantic here...the word "runaway" actually has special meaning in climate science. It is unlikely that Earth can achieve "runaway" greenhouse warming in the strictest sense. This is due primarily because the Simpson-Nakajima limit (and related Komabayashi-Ingersoll limit) on Earth is sufficiently high and because the primary feedback driver would be water vapor which is a condensing gas. The SN limit is the clamp on the outgoing longwave radiation (OLR) in a saturated atmosphere. The SN limit is thus the point at which absorbed solar radiation (ASR) must exceed OLR to bootstrap the runaway phase. I believe the SN limit is around 290 W/m^2, but OLR is currently 240 W/m^2 so there is ~50 W/m/2 of buffer before the SN limit is even reached. The consensus seems to be that there aren't enough non-condensing GHGs to get us anywhere close to this SN limit. However, a "moist" greenhouse has a lower limit. The moist greenhouse is characterized by a state at which the Earth's cold trap near the stratosphere can no longer stop the leaching of water vapor (which would typically condense out and drop back to the surface) into the stratosphere where it would then slowly deplete to space. Aside from the obvious fact that the atmosphere would be in a perpetual moist state this scenario would likely have other undesirable consequences like the destruction of the ozone layer and evaporation of the oceans. There is still considerable debate regarding whether Earth can support a moist greenhouse state. Assuming that it can the estimates I've seen of the CO2 required to bootstrap this process could be 10,000 ppm (give or take) given the current solar output. I will add the caveat that there are notable scientists (namely James Hansen) that aren't quite so confident that a "runaway" phase is not achievable. Here is a pretty good summary style publication that explains things better than I can. Goldblatt & Waston 2012: The runaway greenhouse: implications for future climate change, geoengineering and planetary atmospheres

That first paper says the water vapor feedback is strongly positive at about 2 W/m^2 per C. They end with "The existence of a strong and positive water‐vapor feedback means that projected business‐as‐usual greenhouse‐gas emissions over the next century are virtually guaranteed to produce warming of several degrees Celsius." Unfortunately the second paper is paywalled and I cannot find an open copy. I will say that I took the opportunity to read other publications by the lead author Dr. Sullivan. So far I've not seen anything that leads me to believe he doubts the water vapor feedback or GHGs contribution to warming.

UKMET says it may do something as well.

I agree. I think the aerosol hypothesis can be at least partially falsified by the OHC data. If aerosol increases were playing a significant factor we should have observed a decrease in the EEI and thus a reduction in OHC uptake. I think ENSO or other natural cycles better explain the atmospheric warming hiatus. Cryosphere declines also seem (to me) like a reasonable hypothesis for explaining the hiatus as well, but I haven't seen any literature that I can use to support it at the moment.

Right. No disagreement with the uptake of carbon by the ocean. But oceanic heat content (OHC) is a measure of heat uptake; not carbon uptake. It is directly related to the Earth Energy Imbalance (EEI). The EEI is dispatched into the ocean, air, ice, and land. The hiatus period is characterized by a general pause in heat uptake by the air despite heat uptake proceeding in the ocean. ENSO cycles likely played a role in the waning of the transfer of heat to the atmosphere. I was wondering how much of a role the cryosphere played in that as well. The puzzle is that increased aerosol loading brings down the EEI and thus should have reduced the rate of OHC increases. This is what makes me think aerosol loading may not be as important in explaining the hiatus period as some have hypothesized. @donsutherland1 found the following study which estimates that value at 89%. It appears to be one of those comprehensive style studies that attempts to provide the best estimate of EEI with the lowest uncertainty from the vast body of evidence available at the time of publication. Schuckmann et al. 2020: Heat Stored in the Earth System: where does the energy go? There are some big names in the author list and the bibliography at the end is huge. Keeping in mind that I'm but an amateur...this looks like something that will be heavily relied upon in the upcoming IPCC AR6 report.

Aerosol loading probably explains part of the warming hiatus. One thing that puzzles me is that oceanic heat content kept marching upward. It makes me wonder if the typical transfer of heat into the atmosphere waned during the period only to be taken up by the cryosphere. The post 1998 El Nino period was about the time where cryosphere declines became most acute.

That is a reasonable, rational, and well thought out post @skierinvermont It is important for people to understand that while many of us do advocate for AGW in general we're not all blinded by the fact that there are many things regarding the state of climate science that lead to frustratingly large uncertainty envelopes, annoyingly large spreads between prediction and observation, poor explanations of past observations, etc. I'm always open to discussing climate science's shortcomings. I have a list of pet peeves, concerns, and questions myself actually. My only request is that all discussions be done with an evidence centric approach. Claims on either side of the spectrum should be reasoned, measured, and backed by peer reviewed literature, repeatable experimentation, and verifiable observations. Unsubstantiated claims of fraud, manipulation, conspiracy, the general "nuh-uh" class of arguments, and hostility towards science in general are unsatisfying and unconvincing and do little if anything to move the discipline forward. For those contrarians out there...don't hear what I'm not saying. I'm not saying that criticism isn't welcome. What I'm saying is that if that criticism does not provide a means or path by which our models of the climate system can improve then what good is it really? If you ask me to abandon a model, which I fully understand isn't perfect, for no model at all then I'm going to push back. But if you instead accompany your criticism or insight with a statement of "here's how your model could be better" then I'll be all over it.

How much stronger? How much slower? How much wetter? Those are tough questions to answer on an individual storm basis. I will say that the Siberian heat wave this summer was made vastly more likely by global warming than it would have been otherwise. The shifting of the bell curve and skewing of the probabilities was extreme enough in this particular case that it's hard to claim that the event was merely a fluke. I'm just saying in general it's a stretch to attribute all of an individual event's magnitude on global warming. Even if Harvey were a bit faster or had slightly less water content to work with Houston still would have experienced catastrophic flooding. In that regard I don't think global warming tipped the scale that caused Harvey to transition from a nuance to a catastrophe. But it may have shortened the recurrence interval between storms of similar magnitude.

No. We can quantify its radiative forcing via the well known 5.35 * ln(Ct/Cr) formula where Ct is the target concentration and Cr is the reference concentration. The difference between the Maunder Minimum and Modern Maximum is +0.25 W/m^2 (see Kopp 2016 and data). 2xCO2 comes out to 5.35 * ln(2) = +3.7 W/m^2 (see Myhre 1998). It is about an order of magnitude stronger than the solar forcing. For the Quaternary Period...mostly yes. But it is complicated and most of the lag claims are based solely on Antarctica (see Shakun 2012). For other eras like the hyperthermals...no. The Paleo-Eocene Thermal Maximum is an era in which CO2 catalyzed increases in temperature. The PETM is considered to be the best analog to the contemporary era. Yes. But these are terms that had always been meant to represent cool and warm eras in the North Atlantic region. They were never meant to be applied globally (see Lamb 1982). Globally temperatures have been relatively stable during the Holocene and especially the last 2000 years (see Kaufman 2020). No. Of course not. But neither do I believe that the position he advocates for aligns with the body of evidence either. The warming is likely unprecedented for the Holocene. Meh...it's complicated. Global warming will certainly shift the bell curve of extreme events to the right, but to claim that any one particular event is wholly attributable to global warming is a bigger stretch than most scientists are willing to make in the vast majority of cases.

TD21 is now Tropical Storm Vicky.

Right. I get that you don't accept the evidence that is contrary to the hypothesis...CO2 always lags T, CO2 forcing is small and insignificant, natural forcing is larger than anthroprogenic forcing, positive CO2 feedbacks do not exist, etc. What I'm asking is...what kind of evidence would you accept that would convince you that the above hypothesis are false?

So what kind of evidence would you accept that would falsify the natural hypothesis and convince you that the anthroprogenic hypothesis can survive falsification?

I had not read through that publication yet. Thanks. It is from the University of Washington in Seattle. It is an independent review. Summarizing the trends for the mid-troposphere in the tropical region... as-is tls-corrected UW-obs 0.115 0.160 UW-gcm 0.124 0.170 NOAAv3.0 0.105 0.149 RSSv3.3 0.089 0.125 UAHv5.6 0.029 0.064 UW deployed two analysis methods for handling the diurnal bias: obs for observational and gcm for global circulation model. They also corrected for the stratospheric cooling contamination in the tls-corrected column. UAH is clearly the outlier here.

Exactly. Which is why IGRA especially should not be used for climate research and the type of comparison's presented in the graph above.

Right. Okay. So that comes from Dr. Spencer's blog here and references their paper here. Note the following. 1. It is not independent. 2. This is not a comparison of the global surface mean temperature. It is a comparison of the more narrowly focused tropical region at the mid troposphere layer (which is probably contaminated by the cooling stratosphere BTW). 3. The radiosonde dataset is IGRA whose maintainers specifically warn against using for this type of comparison. 4. The reanalysis dataset is ERA. So based on this chart I'm left with the impression that ERA is the next "best" characterization of reality; according Dr. Spencer anyway. And guess what...ERA says the global surface warming trend is +0.19C/decade which is more inline with Berkeley Earth, GISTEMP, NOAA, RSS, etc. So while UAH may (I question the use of IGRA) have superior skill in the mid troposhere layer in the tropical region it is certainly an outlier on the broader and more widely disseminated global mean surface temperature measure. Given Dr. Spencer and Dr. Christy's history of focusing on the tropical mid troposhere I think this makes sense. But the thing is...most of the other datasets are designed to estimate the global surface temperature instead.

The guy brings up good points. Physical parameterization schemes are fraught with assumptions and other problems. Solving partial differential equations is done numerically only with errors that propagate (chaos theory). Model parameters must be approximated and tuned. Everybody who is anybody understands this. Global circulation models will NEVER be perfect; not even remotely. But...just like any other approximation of reality they still perform reasonably well and are quite useful. In fact, GCMs provide a reasonable (dare I say best) match to the contemporary warming era on decadal scales than any other type of model. Why then would anyone chose to abandon a model with demonstrable skill for an inferior model or no model at all? These kinds of arguments that Mr. Browning is making are what I often call "nuh-uh" arguments. Convince people that an imperfect, but useful model of reality should be abandoned in favor of inferior models or no model at all. Sorry, but I and most other scientists are convinced not by "nuh-uh" arguments but by "here's how to do it better" arguments. Show me a model that performs better than what we already have and I'll be all over it. And besides the implication here is that climate sensitivity is only determined via global circulation models. That could not be further from the truth. As I've already pointed out scientists consider many lines of evidence including modeling (radiative transfer models, energy budget models, GCMs, etc.), observational (paleoclimate, instrumental, etc.), and physical laws (thermodynamics, molecular physics, etc.) for determining climate sensitivity. It is possible that Mr. Browning is unfamiliar with the state of climate science in this regard. I don't know.

This is a challenge other scientists and myself must accept as well. Here is how I select the temperature dataset that best characterizes reality. In lieu of having any compelling reason to select a single representation of reality I equally weight and blend all of the representations available into an ensemble and use the mean as the "best guess" at reality. I take the month-to-month temperature anomalies from as many datasets as are available and let that be my baseline for comparison. When you do this you will see that UAH deviates significantly from the ensemble mean; more so then any other dataset. It is an outlier in most respects. To claim that UAH is the best is to claim that UAH has a monopoly on correctness and that all of the other datasets using wildly different techniques and available input data somehow managed to mistakenly come up with the same wrong answer. How likely do you think that scenario is? I'll answer that...not very likely. And I'd miss a great opportunity for discussion if I didn't point out that many weather forecasters use this same basic approach to make weather forecasts. I have no idea which model will best characterize the state of the atmosphere at some random point in the future. Sometimes the GFS does better. Sometimes the ECMWF does better. Sometimes the UKMET does better. How does one decide which to choose? Well...we don't. We blend them together into a model of models. You've probably noticed that the NHC heavily weights their official forecast on the TVCN and IVCN ensembles because historically the ensemble mean of many models performs better than any individual model alone. In fact, it is my understanding that the human element at the NHC was actually making forecasts worse (OFCL performed worse than TVCN and IVCN) and so the NHC forecasters were encouraged to essentially carbon copy the TVCN and IVCN ensemble means for track and intensities forecasts going forward. It was a lesson in humility that adding human intuition and feeling to the forecasts only made things worse.

I don't see how you can say UAH is the best. What dataset/model are you comparing UAH to to assess its "bestnest". And why did you select that dataset/model for comparison to begin with? Why not just call that chosen dataset/model the best? RSS used to use a GCM to make diurnal bias corrections. The academic community criticized them for it. They changed their methodology in this regard in v4. The warming trend went up. Perhaps the GCM method was more correct afterall? (see Mears 2017). Karl did not adjust SSTs upward. He gets SSTs from ERSST (see Karl 2015). I read the ERSSTv4 papers (see Haung 2015 part 1 and part 2 and supplemental). Now, understanding that I'm not an expert, I did not see any adjustments documented that I felt were mistakes. In fact, quite the opposite. I think it would be a mistake to omit these adjustments and not publish v4 of ERSST. BTW...I believe ERSST is now up to v5. And GISTEMP and others also uses ERSST as well. I don't know about the early 1800's but at least since the 1880's the unadjusted data show MORE warming; not less. (see figure 2B Karl 2015). And again...show me a dataset/model that you feel best characterizes reality so that we can make objective comparisons between it and UAH (or any dataset really). Justify why you think that chosen dataset/model best characterizes reality.

I have to be pedantic here...the cause of the +0.87 W/m^2 EEI is the net effect of ALL radiative forcing agents. CO2 is but one component. What you actually mean is...the dominating factor in the current +0.87 W/m^2 EEI is CO2. Though I must point out that other GHGs like CH4, N2O, O3, CFCs have significant components as well. There are some components for which the uncertainty envelope on their contribution is frustratingly large. I'll throw out clouds and aerosols as examples here.

I trust models to predict the location and timing of solar eclipses. I trust models to predict the behavior of drugs in my body. I trust models to predict how solutions to engineering problems will behave before those solutions are implemented so that I don't waste time and money. I trust models to produce temperature readings from RTDs and thermocouples. I trust models to produce satellite images of clouds and water vapor. I trust models to forecast hurricane track and intensities several days out. I trust models to forecast severe weather outbreaks. Everyone trusts scientific models and even bets their lives on them on a daily basis. We calibrate our trust based on the ability of the model to explain and predict observations. Invoking a model is not something to be ashamed of. It is something to embrace because if you aren't invoking a scientific model then you're just guessing.