2023    87   98  120  100   94  108  119  119  148  134  143  137    117 112     88  105  115  141  2023
Year   Jan  Feb  Mar  Apr  May  Jun  Jul  Aug  Sep  Oct  Nov  Dec    J-D D-N    DJF  MAM  JJA  SON  Year

https://data.giss.nasa.gov/gistemp/tabledata_v4/GLB.Ts+dSST.txt
1.17 so resolves 100%

For those who want to bet more aggressively on the anomaly being quite high (1.17+) or not that high, I made a similar, alternative market. https://manifold.markets/aenews/will-the-2023-nasa-gistemp-temperat

Will the 2023 NASA GISTEMP Temperature Anomaly be 1.17+?

50% chance. This market will resolve positively if the unsmoothed 2023 Global Land-Ocean Temperature Index (1951-1980 Base Period) according to NASA/GISTEMP is 1.17 or higher (can be found here). Only the initial reported value for 2023 will be considered for this market, and any future revisions will not impact market resolution. If that specific page does not update, other reporting from NASA/GISTEMP will be used instead (e.g. here or here). Another Similar Market: https://manifold.markets/ChristopherRandles/pays-10-for-each-001c-2023-temperat [link preview]

October at

https://manifold.markets/ChristopherRandles/pays-2-for-each-001c-october-2023-t

Pays 2% for each 0.01C October 2023 temperature anomaly is above 1.00C per Gistemp versus 1951-1980 base period

50% chance. Data in no smoothing column at https://data.giss.nasa.gov/gistemp/tabledata_v4/GLB.Ts+dSST.csv or https://data.giss.nasa.gov/gistemp/tabledata_v4/GLB.Ts+dSST.txt (or such updated location for this Gistemp v4 LOTI data) Resolution If the anomaly versus 1951 to 1980 base period per gistemp LOTI is reported as 1.50C or higher this claim pays 100% If the anomaly versus 1951 to 1980 base period per gistemp LOTI is reported as 1.00C or lower this claim pays 0% If 1.01C claims pays 2% If 1.02C claims pays 4% and so on.

Clever setup

After August data, Berkeley Earth's projection looks maybe 0.13 degrees higher than the previous warmest year on record, which is 1.02 in GISTEMP, so taking that at face value would give an estimate of 1.15, i.e. a full payout with some uncertainty around it. Berkeley Earth mentioned some concerns about the past months being unusual, which might make their model produce an overestimate, and I think Berkeley Earth is also expected to measure a slightly larger gap between 2023 and 2016 than GISTEMP. On the other hand, September has been extremely hot so far. On the whole, the market looks undervalued to me.

@StevenK I see extremely large discrepancies with Berkley earth's [I meant climate reanalyzer] data and the NASA data (up to .075C for a given year). anyone else analyzed this? Maybe I'm doing something wrong.

@Will58c3 Berkeley Earth has lower estimates for 19th century temperatures, so while differences between datasets don't matter that much for "warmest on record" type questions, they matter a lot for "temperature increase relative to pre-industrial" type questions; see https://www.theclimatebrink.com/p/will-2023-be-the-first-year-above

Will 2023 be the first year above 1.5C?

It is possible, but only in one of the four temperature records that stretch back to the 1800s

@StevenK I don't think that's quite right. If you look at NASA's data they report 2016 was the same temperature as 2022. If you look at the climate reanalyzer data they find a .071 C degree difference with 2016 being the hotter year. But you can find the GISTEMPs estimates for prior months in 2023.

@Will58c3 A diff of 0.071 is within the margin of error of about 0.1C so such a diff is to be expected.

@ChristopherRandles Sure, but it's still a very significant differences for purposes of this market, right? The warmest year markets are done though imo.

Sorry but I don't quite understand how this market works. You can resolve it to a percentage between 0 and 100%?

@Will58c3 I see, I misinterpreted you as referring to Berkeley Earth giving higher anomalies than other sources. Where does the reanalyzer give annual data? I was going by the first graph here, which shows different datasets as having nearly the same relative temperatures of different years, up to about 0.01 degrees or sometimes 0.02.

@StevenK my bad I i meant to refer to climate reanalyzer all along. you can download the climate reanalyzer data as a json file and import into excel.

@Will58c3 This market works entirly on gistemp anomaly. If the anomaly for the year is 1.06 then the claim resolves at 10%.

If the anomaly for the year is 1.07 then it resolves at 20%

and so on.

So far through July & August the anomaly is 1.03 and 1.06 but with really high recent temperatures and El Nino this is likely to continue to increase.

Bringing Berkeley temperatures or climate reanalyzer satellite temperatures into it may be more of a distraction and add too much uncertain differences into it so that such analysis isn't very useful or maybe it still gives a edge for betting purposes. Up to you to decide.

@ChristopherRandles I think Berkeley Earth's projection is still very informative even though it needs to be adjusted a small amount.

@Will58c3 I see - yes, this page does show a much larger difference between 2016 and 2020, and I don't know what to make of that. (edit: sorry, I was looking at 2019, not 2020; but it turns out you said 2022, so I'm confused, because NASA shows 2016 as much warmer than 2022, and the reanalyzer doesn't seem to have data for 2020 or 2022)

@ChristopherRandles I just went down the rabbit hole for the hottest month ever market

Isn't the current anomaly 1.06 C or =(1.24+1.19+1.08+0.93+1+1.2+0.97+0.87)/8 ? https://data.giss.nasa.gov/gistemp/graphs_v4/graph_data/Monthly_Mean_Global_Surface_Temperature/graph.txt

@Will58c3 Yes sorry that 1.03 was through July.

@StevenK ugh my bad i meant 2020 (actually did mean 2022)

@Will58c3 I did not expect such a gap. Interestingly for comparing the difference between 2023 v 2016, the climate reanalyzer data (CRF) and NASA data (GISTEMP) are fairly close..

@Will58c3 They have annual data here, but 2016-2020 looks like a difference of 0.007.

@StevenK Sorry, I need to be more careful and proofread (keep confusing 2020 and 2022). Based on the JSON data from I'm finding a .05C difference between 2020 and 2016 average temp (see https://climatereanalyzer.org/clim/t2_daily/?dm_id=world ). I'm using that data because to my knowledge it's the only one that updates daily.

The annual data you're looking at seems different and based on a reanalysis.

Does "pays" mean "resolves to"?

I am confused, Because my recent market uses the literal meaning of paying.

@KongoLandwalker yes means resolves to. I'll change it

Hope that is OK (don't really want to use resolve in title)