Global Land and Ocean Temperature Records

Ok,

This is a 'work in progress' but it will show the problems and solutions:

Excel Linear Average of Berkely Temperature and Sea Level

Here I was looking at the excel 'running average' trend line for both sea level and Berkeley Earth data. I was investigating the 'pause' that obviously disappears when you draw trend lines of the maximum and minimum. However, if one uses a straight-line average, one can 'invent' a false pause by cherry picking the end points:

This chart show that linear, running averages also have a regional bias. So Texas is 'flat' but Alaska shows a clear upward trend. This pattern is consistent with the models that indicate the greatest and earliest increases will be at the poles and less along the equator. If nothing else, the equator tends to have a more 'even' solar radiance versus the poles that vacillate between 24 hour dark and 24 hour days during the summer and winter seasons.

Excel running average vs Gaussian trend-lines on Sea Level data


Now this chart shows a latent defect I've seen for years in excel trendlines:

• red line - an excel trend line for the Jason sea level data. Easy to generate but wrong!
• triple blue line - an excel function generated, trend line . . . I coded that average and know it is accurate!

I was stung by this latent excel bug in the past so when I'm using any sort of linear average and I want it to be true, I generate my own average range function in a column and plot it. A little more work but I know it is true and can be audited. But notice how the sea level data has a pattern of spikes.

It was obvious there is an annual cycle in the sea level data by looking at the peaks and valleys. So I decided to use an annual period, Gaussian filter:

Now we can see local events like the 1997-98 jump in sea level and the 2011 sag. These are not evident in the earlier plots. Now a running average with a one year span would see them but trust me, their magnitude is suppressed. It has to do with how a linear average applies a constant weighting to each point in the linear sequence. If you are aware of this, dealing with linear averages holds no terror but try to explain it . . . very difficult.

Gaussian weighted average on Berkeley Earth data


vs

• Up until the 1998 El Nino, we see a faster rise in the Berkeley Earth reported, global temperature rise.
• After the 1998 El Nino, the rise continues as evident in the peaks and valleys but at a somewhat slower pace.
• Sea level rise continued in a somewhat linear fashion with peak around the 1998 El Nino and valley around the Icelandic volcano and a 'weak' La Nina.
  
• The sea covers 71% of the earth's surface, a super-majority of the part exposed to planetary warming.

My thinking is land based temperatures are likely to be impacted by the lower, thermal properties compared to sea water. Land temperatures appear to be 'noisier' than sea level. So if I can come up with a credible way to read out the temperature effects versus land, ice melting, we'll have a true, global thermometer. Right now, it appears land ice melting and snow deposits are the major signal.

Now the difference between years of sea level change, minus the thermal effects (if I have any success), means we can calculate the mass of melted and/or drained land water. So this is the type of problem that interests me.

Austin and I had discussed the 'merit' of a global average temperature in the past versus regional effects. Global average has the advantage of 'easy to understand' but sadly doesn't really tell us about the specifics. It was part of a discussion about the 'Little Ice Age'. There is another paleo-record event of equal interest to me, "The Year Without A Summer", 1816. But that goes into a subject called climate forcing functions which interests me but leads many others to have eyes that glaze over before they gently flutter closed. <grins>

Bob Wilson

 

Bob, what is the Excel trend line function. Spline? Must be documented somewhere

As this thread was carefully named it would not include RSS and UAH troposphere microwave retrievals. However those should be presented somewhere.

 

Well we can agree the word begins with "S". <GRINS>

Every now and then I think about trying to figure it out but . . .

I know, perhaps someone who is an advocate of those sources might 'pickup the burden' and bring it here. But I'm not interested in reading a 'second hand' link to some dubious author. If they don't have the confidence and skills to bring their own work, I have no time to waste on copy-and-waste.

Bob Wilson

 

Tell me how you know that sea temps had any accuracy prior to Argo.
Science fiction ,is what you are presenting.
In other words you are presenting BS.

Nice scientific looking colored graphs though.Look totally legitimate except the data is BS.

 

Over 24 hours later and still nothing.

Bob Wilson

 

2015 September T must be 'out', with October coming soon. Finally, this year will be hot, and whether it becomes 'the hottest' will depend on what data and statistics you prefer.

The last few months of 2015 were certainly affected by El Nino positive. I think this put folks asserting not warming in a bit of a box. It would be difficult to treat 1998 and 2015 (both El Nino positive years) differently. All recent decades have become warmer, and marine processes including ENSO seem to have become less effective at holding that back.

It may be too soon to relegate 'global cooling any time now' folks to the dustbin, but they are much in need of a miracle. We might instead look at recent decadal trends and take those into the near future. I have already made my call that the first half of this century will be warmer, but not painfully so.

The second half century, and the one following, could challenge the human enterprise. Lots has been written about that, but only our youngest readers here will have a direct personal stake. Best of luck to them!

 

UAH and RSS satellite look-down T analyses were high in 1998, so I'd expect them to jump up soon. How to spin that?

 

Funny seeing this:

Earlier this evening I checked the Unisys SST:

That is one impressive El Nino. Of course there is no relationship to climate. So El Nino must be a weather?

Roll of the dice, there has been an impressive Pacific hurricane last week that even after passing over the mountains of Mexico flooded a lot of East Texas. The left-overs moved over Louisiana and Mississippi and are now moving into North Alabama. We'll have to wait 30 years before we can call it a 'climate.'

Bob Wilson

 

Very recently published

Industrial-era global ocean heat uptake doubles in recent decades
Peter J. Gleckler, Paul J. Durack, Ronald J. Stouffer, Gregory C. Johnson and Chris E. Forest
Nature Climate Change
DOI: 10.1038/NCLIMATE2915

They do it all, reanalysis of 19th century data*, recent cruises, recent ARGO data, modeling. Sort of a 1-stop shop.

*It is correct to consider old ocean T data as sparse, and that seems like a generous description. But one needs to take that up with the authors, and other practitioners in the field.

Similar topic, Jason 3 is now ‘up there’, to measure sea level and other interesting things.

 

Because BobW (our sea-level guy) is not keeping that topic on our first page, I'll drop this in here:

Revisiting the contemporary sea-level budget on global and regional scales
Roelof Rietbroek, Sandra-Esther Brunnabend, Jürgen Kusche, Jens Schröter, and Christoph Dahle
Proc. Nat. Acad. Sci. USA, in press, doi: 10.1073/pnas.1519132113

They have refigured ongoing thermal expansion and some other things. Read, me hearties, or be known for not doing so.

 

Thanks!

There is a lag between data being available so I've been a little slow. But I'm leaning towards shifting my 'annual' update to the Arctic and Antarctic sea ice minimums plus a three week delay for getting the data. The reason is these are empirical data points and the extreme end-points. However, I could be persuaded to do the mid-point between these two or four updates per year.

Choice,
Bob Wilson

 

There is an earthly environment where T varies from -10 oC in January to -8 oC in July. You might want to live there if you like cold, but food would have to come from somewhere else. Its main ‘claim to fame’ is being the place (the only place) where T has NOT increased since a notable upwards bump in 1998.

It is global TLT defined by microwave emissions as seen by satellites and transformed to T by UAH and RSS. Bugging you with this now because I did not know T or seasonal variation there before. Spencer’s UAH website gives us this now (big thanks). Before we have only seen TLT anomalies based on a prior index interval.

It’s an abstraction; not really a place. More like above everywhere, at undefined altitudes. It is notable for being DIFFERENT from the earth’s surface, where T has increased since 1998. Therefore people talk about it A LOT, even though they can’t say where it is. Harsh to call it a mythical unicorn

Here on the surface, T is increasing, matters not which T compilation you favor. Matters A LOT to presume that all of them are running the same +T hoax. No hoax could persist with so many eyes looking at T, but I admit that unicornists might suppose otherwise.

Don’t like current surface T trends? Go live at TLT. Bring warm clothes, and lots of snacks.

 

ROTFLMAO.

Bob Wilson

 

2016 Feb. UAH TLT went very high, exceeding 1998. Such a bother. do not know where to look now to support claims that global warming has stopped.

I have $1000 ready to bet against decadal global cooling. Details would have to be worked out.

 

Source: http://nsstc.uah.edu/climate/2016/january/jan2016GTR.pdf

Bob Wilson

 

I know we try to avoid referencing an advocacy site but this showed up: Global Warming Basics: What Has Changed?


















Since the data charts are so easily collected . . . carpe diem.

Bob Wilson