bdgwx

This the Osman et al. 2021 publication. A non-paywalled version can be found here. It's yet another hockey-stick graph. As I often tell others we have a whole hockey league of hockey-stick graphs now. They come out so often now that it's nearly impossible to keep track of them all anymore.

I thought you all might be interested in this. I created the following plot using 4 traditional, 2 satellite, 1 raob, and 1 reanalysis datasets of the global mean temperature including an equal weighted 3 month centered composite of the 8 datasets and compare everything with the CMIP5 ensemble mean from 1979/01 to 2021/08. All timeseries are using the same baseline for comparative purposes. The composite trend is +0.19 C/decade.

The NSIDC site now has an http link for their data. This is useful since many browsers no longer support FTP. https://masie_web.apps.nsidc.org/pub/DATASETS/NOAA/G02135/seaice_analysis/

Per the official 12Z NHC wind profiles the IKE is 33 TJ. Per the RAMMB profiles the IKE is the 80 TJ. I'm a little suspicious of the NHC wind profiles but if we split the difference that would give us 56 TJ. This is a pretty big deal.

Using the wind radii from the 3Z NHC advisory with VMAX = 90 kts and RMW = 10 nm the IKE has increased to 32 TJ.

Per the official 21Z update the IKE is now 23 TJ.

** 2021 ATLANTIC RI INDEX AL092021 IDA 08/28/21 18 UTC ** (SHIPS-RII PREDICTOR TABLE for 30 KT OR MORE MAXIMUM WIND INCREASE IN NEXT 24-h) Predictor Value RI Predictor Range Scaled Value(0-1) % Contribution 12 HR PERSISTENCE (KT) : 20.0 -49.5 to 33.0 0.84 14.8 850-200 MB SHEAR (KT) : 10.7 30.1 to 2.3 0.70 5.7 HEAT CONTENT (KJ/CM2) : 88.0 0.0 to 151.8 0.58 4.3 STD DEV OF IR BR TEMP : 8.2 36.6 to 2.8 0.84 7.0 MAXIMUM WIND (KT) : 90.0 22.5 to 137.5 0.62 4.2 BL DRY-AIR FLUX (W/M2) : 212.2 895.4 to -55.0 0.72 4.8 2nd PC OF IR BR TEMP : 0.3 2.9 to -2.9 0.45 2.2 POT = MPI-VMAX (KT) : 67.3 28.3 to 146.3 0.33 1.0 D200 (10**7s-1) : 31.0 -29.7 to 185.9 0.28 0.8 %area of TPW <45 mm upshear : 0.0 100.0 to 0.0 1.00 0.6 SHIPS Prob RI for 20kt/ 12hr RI threshold= 52% is 10.7 times climatological mean ( 4.9%) SHIPS Prob RI for 25kt/ 24hr RI threshold= 56% is 5.2 times climatological mean (10.9%) SHIPS Prob RI for 30kt/ 24hr RI threshold= 46% is 6.7 times climatological mean ( 6.8%) SHIPS Prob RI for 35kt/ 24hr RI threshold= 39% is 10.0 times climatological mean ( 3.9%) SHIPS Prob RI for 40kt/ 24hr RI threshold= 30% is 12.5 times climatological mean ( 2.4%) SHIPS Prob RI for 45kt/ 36hr RI threshold= 21% is 4.5 times climatological mean ( 4.6%) SHIPS Prob RI for 55kt/ 48hr RI threshold= 12% is 2.6 times climatological mean ( 4.7%) SHIPS Prob RI for 65kt/ 72hr RI threshold= 0% is 0.0 times climatological mean ( 5.3%) Matrix of RI probabilities ------------------------------------------------------------------------------ RI (kt / h) | 20/12 | 25/24 | 30/24 | 35/24 | 40/24 | 45/36 | 55/48 |65/72 ------------------------------------------------------------------------------ SHIPS-RII: 52.3% 56.1% 45.6% 39.1% 29.9% 20.6% 12.2% 0.0% Logistic: 26.0% 42.2% 32.2% 17.4% 8.8% 12.1% 9.3% 1.2% Bayesian: 26.0% 55.7% 25.6% 53.5% 15.3% 5.7% 0.1% 0.0% Consensus: 34.8% 51.4% 34.4% 36.7% 18.0% 12.8% 7.2% 0.4% DTOPS: 46.0% 21.0% 19.0% 15.0% 2.0% 0.0% 0.0% 0.0%

Ultimately it boils down to the fact that station moves, instrument changes, time-of-observation changes, etc. all contaminate the record with known biases. These biases must be addressed. It is that simple. It is unethical at best to use the unadjusted data when reporting climatic information like the warming rate, heat waves, etc. given these documented biases especially if you have been notified of the existence of the issue.

Here is a great post by Gavin Schmidt on the ECS. https://www.realclimate.org/index.php/archives/2021/08/notallmodels/

It appears that the IPCC has adopted the phrase "practically ice-free" to mean the same thing as what we commonly use of "ice-free" to mean < 1e6 km^2 of extent. B.2.5 Additional warming is projected to further amplify permafrost thawing, and loss of seasonal snow cover, of land ice and of Arctic sea ice (high confidence). The Arctic is likely to be practically sea ice free in September at least once before 2050 under the five illustrative scenarios considered in this report, with more frequent occurrences for higher warming levels. There is low confidence in the projected decrease of Antarctic sea ice. AR6 WG1 SPM pg. 20. That is the prediction...2050 with a 66% likelihood.

Correct. Note that "likely" means 66% likelihood and "very likely" means 90% likelihood. The "very likely" range is 2.0-5.0C. The "likely" range is 2.5-4.0C. This is the first report that provides a best guess...3C. The IPPC divides the assessment reports into 3 parts called working groups. WG1 is the Physical Science Basis, WG2 is Impacts, Adaptation, and Vulnerability, and WG3 is Mitigation of Climate Change. It may be the WG2 report that is expected to be released in March '22.

This will take months to read, but the headline topic of equilibrium climate sensitivity comes as a surprise to me. I was not expecting the big jump from 1.5 to 2.5C on the lower bound. A.4.4 The equilibrium climate sensitivity is an important quantity used to estimate how the climate responds to radiative forcing. Based on multiple lines of evidence, the very likely range of equilibrium climate sensitivity is between 2°C (high confidence) and 5°C (medium confidence). The AR6 assessed best estimate is 3°C with a likely range of 2.5°C to 4°C (high confidence), compared to 1.5°C to 4.5°C in AR5, which did not provide a best estimate. SPM pg. 14

If 8/6 ends up being the minimum that would be epic indeed.

The long awaited Assessment Report 6 from the IPCC Working Group 1 regarding the Physical Science Basis Report has been released. https://www.ipcc.ch/report/ar6/wg1/

Some sea ice coupled GCMs did show that the 2020's could be a stall decade, but definitely not a recovery. I don't even know how a recovery could be possible considering the planetary energy imbalance is sitting at close to +0.9 W/m2 right now. And it might even be increasing at that. And given that environmental conditions this year should have promoted higher sea ice extent/area and yet we're still well below average is rather telling. Recovery is a pipe dream.

RSS matches RATPAC better. https://www.ncdc.noaa.gov/sotc/upper-air/201813 The claim that UAH matches radiosondes better comes from Dr. Spencer and Dr. Christy themselves. They also use IGRA as their radiosonde data source. The IGRA website warms users against using their dataset for long term trend analysis. How can you know that UAH (or any dataset for that matter) is "true" and thus a gold standard?

One thing that confused me a couple of years ago is the increase in outgoing longwave radiation (OLR). I naively thought that GHGs would reduce OLR and that's what creates the Earth Energy Imbalance (EEI). That is what happens initially. But...the warming that results from a positive EEI leads to feedbacks like the cloud feedback. If the cloud feedback is positive then Earth's albedo will drop and the absorbed shortwave radiation (ASR) will also increase. Because EEI = OLR - ASR there two ways for the planet to achieve energy balance after a positive EEI perturbation. The first is if OLR increases (temperature increase). The second is if ASR decreases (albedo increase). The fact that we observe an increase in OLR while EEI itself continues to increases is a tell that ASR is increasing which means albedo is decreasing. This is consistent with the hypothesis that the cloud feedback is positive. Donohoe et al 2014 has a pretty good explanation for the counter-intuitive OLR increase when the climate system is acted on by increases in GHG. See figure 1c and 1d for how the OLR and ASR respond to pulses of GHG forcing. It is interesting to note that most (not all) global circulation models actually predict this behavior. DeWitte & Clerbaux 2018 point out that OLR is indeed increasing and attribute this to "cloud thinning".

The low was 104 and the high was 130. I think that might qualify as the warmest day ever observed anywhere in the world.

It's still above 2010, 2012, and 2016. But it is below 2017, 2018, 2019, and 2020.

Perfect. Thanks. Excluding 2021 the mean was 104F (40C) with a standard deviation of 3.0F (1.7C). The z-score for 121F (49.6C) is thus 5.6 implying a recurrence interval of 1-in-10,000,000 years. Including 2021 the mean is 105F (40.5C) with a standard deviation of 5.1F (2.8C). The z-score for 121F (49.6C) is thus 3.1 implying a recurrence interval of 1-in-1200 years. Clearly this was an astonishingly rare event in statistical terms.

While we were all distracted by the historic Pacific Northwest Heatwave of 2021 the 5-day daily extent in the NH dropped below 2020 for this date. It is now the 4th lowest for this date.

So the old Canadian record was 113F (45.0C) in 1937 and the new record is 121F (49.6C) in 2021. Let's say the recurrence interval on the 113F is 50 years and the mean annual Tmax is say 102F with a standard deviation of 5.4F for an annualized z-score of 2.0 (1-in-50 years). Using the mean of 102 +/- 5.4F (1-sigma) the annualized z-score on the 121F would be 3.5 (1-in-5,000 years). Obviously that was just a back-of-the-envelope estimation without any hard data. I have no idea how far off that 102 +/- 5.4F figure is; could be a lot. If someone can supply the annual Canadian Tmax values we can compute the real mean and standard deviation and then just plug all of that into a z-score calculator and get the real recurrence interval for the 121F. My bet...the real recurrence interval from rigorous statistical analysis will be at least several hundred years and probably over a thousand and maybe even approaching the 5000 year figure I guesstimated above.

William Reid is partnering with Chris Burt on this. Reid's blog has a lot more information and is up-to-date through 2020. It is my understanding that they are (or were) preparing a publication to be submitted to WMO. I am unsure of the status on that. I do know this isn't the first time Burt has gotten a record dismissed by the WMO so he's no stranger to the process. http://stormbruiser.com/chase/2020/11/23/death-valleys-134f-record-temperature-study-part-one/

I hope we see a statistical analysis similar to the one we got for the 2020 Siberian Heatwave after this is all said and done. I'd really like to know how this stacks up statistically. I did see the tweet above regarding 500mb heights approaching a 5-sigma event implying a recurrence interval of over 4000 years. Clearly this is an unusual event.