Winter rainfall 2015/16, Hadley series

Posted: April 7, 2016 by tchannon in Analysis, Natural Variation, Uncertainty, weather

Tim writes,

Image

Figure 1, Met Office Hadley monthly rainfall series for England and Wales start date 1766. Winter 2015/16 was wet but ordinary. (data processing by the author, see previous articles)
Plots for all data series as PDF (2MB).

Some other parts / regions of the country do show an extreme but this adds weight to the flicker noise (or 1/f) hypothosis for weather noise.

If independent flicker noise sources are averaged or say weather over a large area are averaged the result tends to Gaussian noise. This for weather agrees with the hypothosis.

Similarly flicker noise tends to local extremes, as we see rainfall extremes, tend to very local and rare for an exact location.

Areal datasets

The 1910 Met Office areal data is also available plotted for March. Can now see the winter spike as a spike.

Winter[1] 2015/16 in most cases has as strong positive correlation between rainfall and temperature, a nice instance of warm wet south westerly airflow smothering cold and dry, hence the warm wet winter.
1. Some months during meteorological winter

Image

Figure 2, an illustrative example using one region of the warm /wet and cold /dry effect in the data. In this case December 1934 is an extreme.

Image

Figure 3, from Meteorological Office report 1935 for December 1934.

Link to PDF at Met office

Areal plots
Tmin
Precipitation
Sunshine
Tmax

Tmean

If anyone wants to explore the data, such as for other regions, I can supply the files. (would be text csv in a zip)

Post by Tim,

I’m briefly around.

Comments
  1. erl happ says:

    Hi Tim, Here is a theoretical framework to explain natural climate variation. It fits your comment “strong positive correlation between rainfall and temperature, a nice instance of warm wet south westerly airflow smothering cold and dry, hence the warm wet winter.”

    I refer to reanalysis data here: http://www.esrl.noaa.gov/psd/cgi-bin/data/timeseries/timeseries1.pl
    Looking at the latitude band 30-60°north temperature is most volatile in the months of January and February. Peak was experienced at the turn of the century and a decline set in from that point: https://reality348.files.wordpress.com/2015/12/air-t-30-60n1.jpg 2016 data not included.
    January and February temperatures are tied to the evolution of the Arctic stratosphere. In fact all latitudes north of 30° south experience the greatest range in temperature in January and February.
    South of 30° south, volatility in temperature is greatest in June and July, tied to Antarctic stratospheric processes. Data here:https://reality348.wordpress.com/2016/01/15/8-volatility-in-temperature/
    Basically we are looking at the northern and southern annular modes of inter-annual and inter-decadal climate variation involving shifts in atmospheric mass from high to other latitudes that alters albedo and the intensity of the planetary winds. In the mid latitudes of the northern hemisphere it also involves a swing between the incidence of warm moist winds from the south and cold dry winds from the north. So the swing is much greater in the northern than in the southern hemisphere where, in the mid and high latitudes the westerlies that originate in the high pressure cells of low latitudes are always dominant. Massively low surface pressure prevails on the margins of Antarctica, albeit varying on centennial time scales, the central feature of the natural climate change phenomenon.
    Greatest volatility in temperature is experienced closest to the source of that volatility in very high latitudes. It is there that the interaction between stratospheric air rich in ozone and mesospheric air that is ozone poor determines the intensity of polar cyclones that are collectively responsible for the shifts in atmospheric mass. Mesospheric air descends in winter when surface pressure in high latitudes peaks. So, temperature volatility is very much a winter phenomenon.

  2. ntesdorf says:

    Nothing to see here folks, move along now. Let’s all get back to that old fashioned ‘Global Warming’.

  3. tchannon says:

    You are comprehensive Erl.

    In the UK variation during the year tends to small, the maritime effect.

  4. erl happ says:

    Tim,
    Where I live in Western Australia the maritime effect is also strong. When I discovered that the range of variation in temperature is very different according to the month of the year it was a eureka moment. Its the signature of the origin of climate change that’s written into the temperature record.

  5. Brett Keane says:

    Erl, is the ” Massively low surface pressure prevails on the margins of Antarctica” because of the sea/ice T difference, or otherwise?

  6. erl happ says:

    In reply to Brett. Thanks for the question. What I write is new information. You won’t find it in the climate literature. You must evaluate its validity for yourself. It describes the mode of natural climate change…in fact the only game in town.

    To see the atmospheric pressure situation in the ocean off Antarctica compare January with October atmospheric pressure here:http://ds.data.jma.go.jp/gmd/jra/atlas/en/surface_basic.html

    To see the progress of change in surface pressure at 60-70° south over time look here:https://reality348.files.wordpress.com/2016/03/slp-60-70s.jpg?w=1496

    Gordon Dobson, reader in meteorology at Cambridge observed in the 1920’s that winds increased in velocity from the surface to the elevation of the tropopause and fall away again above that point. That’s where we find the Jet streams.

    Dobson built an instrument to measure total column ozone. He discovered immediately that total column ozone is greater in low pressure cells than in higher pressure cells. In fact from the edge of a high pressure cell to the core of that cell total column ozone falls off about 25%.

    Ozone absorbs infrared energy from the Earth itself instantly and continuously transferring energy to adjacent molecules. Air containing ozone becomes less dense as it gains kinetic energy of motion.

    Ozone is barely present in air from the mesosphere that descends over Antarctica, especially in winter when surface pressure is high. This air stays cold.

    Ozone is in comparatively low concentration in the air above high pressure cells that frequent low and mid latitudes.

    The largest difference in ozone concentration is found between air that is located above the Antarctic continent and air that is above the ocean on the margins of Antarctic. The is referred to as the Polar Front or the edge of the Polar Vortex. What’s happening? Extreme differences in air density in a horizontal domain from about 8km to 16km result in extreme winds. Winds travel along the isobars and these isobars are very close together. The circulation that is so established is cyclonic, travelling clockwise in the southern hemisphere. The circulation ascends because at its heart is ozone rich air of low density mixing with ozone poor air from off the Antarctic continent and air drawn in from the ‘troposphere’. Air from below the region of strong contrast in air density is sucked in and the result is the lowest surface pressures to be found anywhere on the planet except for that in the core of a tropical cyclone. This is natures equivalent to a vacuum cleaner. The engine driving the circulation is aloft rather than at the surface. We don’t see it, we don’t feel it and being land based animals we naturally think that the surface is the ‘middle kingdom’ at the heart of where everything worthy of our attention happens. No, the birds know differently.

    As the ozone content of the stratosphere changes so too does air density. The motor that drives the vacuum cleaner speeds up or slows down.

    The global circulation varies according to shifts in atmospheric mass that occur most strongly in the winter hemisphere. If those Polar Cyclones intensify it lowers surface pressure over the entire area south of 50° south latitude. Air moves to the mid latitudes and further afield. Shifts in the origin of the wind alters surface temperature. Albedo also changes with the ozone content of the air but that’s a story for another day. Or you can see it here: https://reality348.wordpress.com/2015/12/29/3-how-the-earth-warms-and-cools-naturally/

    Sea ice extension doubles the effective area of Antarctica in winter. That growth of ice area is not accompanied by a shift in the zone of low surface pressure, so the answer is no, surface pressure is not related to the difference in the nature of the icy continent and the icy seas surrounding it. Overall, surface pressure is highest over Antarctica in winter due to the heating of the air in the northern hemisphere and the associated shift in atmospheric mass across the equator. However, surface pressure on the margins of Antarctica is lowest in October when the ozone content of the atmosphere over the ocean is highest and there is an ‘ozone hole’ in the lower stratosphere over Antarctica. But that’s another story again.

    More in this weeks post: https://reality348.wordpress.com/2016/04/08/19-shifts-in-atmospheric-mass-in-response-to-polar-cyclone-activity/

  7. Erl, I am at the other side of Oz. I have 120 yrs of monthly rainfall in a spreadsheet and also daily rainfall but the latter not all accessible in a spreadsheet. The rain is seasonal with the summer months Jan, Feb and Mar accounting for roughly 40% of the annual rainfall. Maximum daily and monthly rainfall in the months Dec to Apr are due to cyclones or remains of cyclones from the east.
    I note that that for each month of the year the rainfall distribution is Poisson where the average rainfall for a month approximates the standard deviation. Note it is not possible to get less that zero rainfall. Maximum rainfall in any month over the 120 years is about 6 *SD. It is possible to see some cyclic pattern especially the high rainfall in the 1890’s followed by the low rainfall in the Federation drought period. Of recent rainfall 2009 to 2013 has been above average 2014 was well below average , 2015 was right on average and the first 3 months of this year well below average (just under 60% of average)
    There maybe some connection between rain and temperature mainly from cloud cover. I do not measure temperature but to me the summer temperatures in south east Queensland have been pleasant and even mild. Warmer than normal sea temperatures ( I think around 24C) have attracted tourists. Day temperatures have rarely this year exceeded 30C.

  8. erl happ says:

    Cementafriend. Nice concept in that name. You are a bit out of my depth. I haven’t looked at Eastern States rainfall. I reckon it would be complex because El Nino events can come from either the northern or southern hemisphere via warming in higher latitudes.

    I would start by plotting surface pressure at 60-70° south against your rainfall. I might then look at a comparison between your rainfall and surface pressure in the south East Pacific off the coast of Chile. Then compare rainfall with surface pressure in the upper regions of the North Pacific in the region of the Kamchatka Peninsula through to Alaska where northern hemisphere ozone and surface pressure falls very low in winter. Finally plot rainfall against surface pressure over Indonesia where evaporation drives the moisture supply that finds its way south east across the Pacific.

    Surface pressure data here: http://www.esrl.noaa.gov/psd/cgi-bin/data/timeseries/timeseries1.pl
    How to download data and get it on to a spreadsheet here: https://reality348.wordpress.com/2015/12/24/2-assessing-climate-change-in-your-own-habitat/

    I am sure you will come up with something of interest that is of more use than the Southern Oscillation Index.

  9. Paul Vaughan says:

    EH advised CF

    “I would start by plotting surface pressure at 60-70° south against your rainfall. I might then look at a comparison between your rainfall and surface pressure in the south East Pacific off the coast of Chile. Then compare rainfall with surface pressure in the upper regions of the North Pacific in the region of the Kamchatka Peninsula through to Alaska where northern hemisphere ozone and surface pressure falls very low in winter. Finally plot rainfall against surface pressure over Indonesia where evaporation drives the moisture supply that finds its way south east across the Pacific.”

    Efficiency Tip:
    You can just drop the rainfall series into KNMI Climate Explorer (it’s user-friendly) and let it color-code a global-scale map of correlations. It will tell you exactly where the strongest correlations are in bright color.

    The Climate Explorer lacks the functionality needed to do more thorough analysis.
    4 things it needs developed:
    1. ability to do spatial derivatives & integrals for fields
    2. ability to do temporal derivatives & integrals for fields
    3. ability to step multiparameter wavelets through above and animate map output (this would just be a batch map-correlation function-call so we’re only talking about adding a do-loop to the code — the user could specify the wavelet vectors, but the platform would have to be able to process multiple series in sequence and output a sequence of numbered correlation maps that the user could assemble into an animation)
    4. ability to define spatial sampling windows with flexible shapes (rather than being constrained to windows parallel to grid) (I recommend designing for mouse-trace boundary input)

    Lack of adequate spatiotemporal analysis tools remains a deal-breaking obstacle in climate exploration. People go on about prediction. The prerequisite spatiotemporal exploration needed to decide how to do spatiotemporal prediction hasn’t even been done yet. The software needed to do it hasn’t been assembled. It’s possible no one has even yet conceived all of the spatiotemporal algorithmic tools needed, never mind secured the funding and time to actually undertake such an ambitious construction project. Certain types of insight will remain roadblocked until it’s done. Climate is spatiotemporal so adequate spatiotemporal tools are needed. What few makeshift partial spatiotemporal analysis tools are available currently are grossly inadequate and yield information about only fractions of the full exploration needed to secure comprehensive vision. Until the needed tools for detailed spatiotemporal analysis are built, attention can still be productively be focused on spatiotemporal attractor fundamentals.

    At least with the functionality of the Climate Explorer tools as they currently exist, CF can have a correlation map in about 20 seconds. It’s just: Copy the time series. Paste it. Click a few buttons. Voila: correlation map. There’s nothing to it.

  10. tchannon says:

    Recent Talkshop readers might be perplexed by Erl Happ. There is a lot of previous reading needed. Erl, an Australian wine grower, has written a lot of good material on large scale climatic processes, earth systems, particularly atmospheric. Look in the left Talkshop sidebar for two links to Erl’s sites, old and new.

  11. erl happ says:

    Paul,That’s an illuminating comment.

    Correlations are what we look for. However, correlations appear and disappear depending upon the period chosen. At least if you graph the two series you can see this correlation appearing and disappearing. Making graphs is also a two or three click process.

    There are other advantages in having your data in a matrix. You can see whether the annual average provides a meaningful relationship or whether the correlation is just seasonal. You can work out the extent of variability in the data by season or by month, a very important clue as to where you should look for the actual cause of variations.

    Correlation is a first step. But, all atmospheric processes are linked and a lack of correlation is no indication that there is not push and shove involved. An excellent example is the relationship between polar surface pressure and the temperature of the stratosphere.Let me explain.

    In winter surface pressure rises across the entire winter hemisphere, over cold land masses and icy northern waters that are in darkness for up to six months. With increased surface pressure over the pole and the proximity of a zone of low surface pressure caused by ozone heating above 8km of elevation in the 60-70° latitude band, always located over the ocean, there is a circulation set up. Air descends from the mesosphere over the pole and feeds into a circulation of ascending air at 60-70° south, particularly obvious in the southern hemisphere because it forms a distinct annular ring about the continent of Antarctica. So, we have descending air over the pole and ascending air at 60-70° north (and south) close to where ozone partial pressure peaks in winter.

    Because mesospheric air has a temperature of -85°C or thereabouts it causes massive cooling that is quite unrelated to surface conditions. The presence of this tongue of air precludes the entry of warmer wetter air flowing polewards from the tropics.

    The flow of mesospheric air is surface pressure dependent. So, we would expect the temperature of the stratosphere to drop as soon as surface pressure increased as it does regularly every winter and from month to month within the winter season. Between 1948 and 1978 that was the case but as surface pressure fell across all southern latitudes over that period the stratosphere warmed. At that point (1978), the temperature of the stratosphere at 10hPa having increased by 50°C in September, the month where the increase was greatest, the relationship between pressure and temperature terminated. Between 1978 and 1998 the temperature of the stratosphere over Antarctica gradually fell indicating a slow recovery in the flow of mesospheric air over the continent. However, the month to month relationship between surface pressure and the temperature of the stratosphere is not evident. After 1998 the relationship resumed.Now, every time surface pressure increases at the pole the temperature of the stratosphere falls.

    With the advantage that a study of monthly data affords over a long period of time we can see relationships come and go and we are in a better position to work out what is happening….modes of causation. That’s absolutely vital.

    What does this change in surface pressure and temperature mean in high southern latitudes? It means that the basic parameters of the climate system are forever changing. We don’t have just one climate system. I’ts on a slider. There are a 100 million climate systems. So, expect your correlations to come and go. Predictability goes out the window.

  12. erl happ says:

    Hey Tim, I would hope that there is not too much reading required. Just stick to the numbered chapters at: https://reality348.wordpress.com/. Ignore the earlier blog. I was learning as I went along. I put the whole thing aside for a couple of years but kept in touch with the research. Learnt more. The started writing about June last year.

    There is also a table of contents appended to the last posts and this will be updated with each post. Just so that its easier to find your way round. Bear in mind that it follows a logical progression. Its a book rather than a blog. I am re-writing as I go along but it had its final shape before December when the first chapter appeared.

  13. Paul Vaughan says:

    Erl, it appears that there has been a serious misunderstanding. The correlations that matter are for the attractors, not the shiny bouncy things.

    The visualization tools I envision go way, way, way beyond what exists. It’s a monstrous programming project. The purpose isn’t to discover shiny bouncy correlations. It’s to discover attractors …around which multivariate things bounce in myriad correlation-crushing manners.

    Indeed we’ve had a serious misunderstanding.

    But we can shake it off and get back on the road.

    : ]

    Cheers

  14. Paul Vaughan says:

    Btw Erl KNMI Climate Explorer does all of the things you recommend to CF.
    It would be helpful if it could additionally map scatterplot-matrices (one scatterplot on each grid cell).

  15. ren says:

    The jet stream to make a circle in the south of the UK.

  16. ren says:

    Click.

  17. erl happ says:

    Paul, Sorry if it appears that I am contradicting what you say in the latter half of your post. Not my intention at all. Its outside my field of expertise. My comment related to the first part reproduced below:

    Efficiency Tip:
    You can just drop the rainfall series into KNMI Climate Explorer (it’s user-friendly) and let it color-code a global-scale map of correlations. It will tell you exactly where the strongest correlations are in bright color.

  18. erl happ says:

    Ren,
    Your diagram illustrates just how unruly the circulation is in the northern hemisphere Big swings in the jet stream are inevitable without a strong anchor of high surface pressure over the Arctic. Secondly there is the bias to the circulation afforded by the consistently low surface pressure associated with high ozone values across the North Pacific. The southern hemisphere is much more orderly.

  19. Paul Vaughan says:

    Erl, clarification: The bright spots on such a map suggest coordinates for which to extract series for graphing. The idea there is not to replace graphing, but rather to let the map suggest places to explore with graphs. Perhaps the exercise will point to exactly the places you suggested to CF. (I suspect you’ve already done enough geographically-referenced graphical exploration that you already know where the map would point.)

    Reassurance: We certainly agree that there’s “push and shove” (in time & space) as you say:


    (from 5 years ago)

    Some peripheral commentary, mostly just to extend and enrich discussion:

    Recall that as Tomas Milanovic has advised us, a full theory of spatiotemporal chaos has not yet been developed. I envision a method that will rip out the turbulence and leave the spatiotemporal attractors. In concept it’s simple. It’s based on spatiotemporal fractional differintegrals. I suspect that a concept as powerful and as simple as central limit theorem awaits proof in the spatiotemporal context. Presently I’m not able to formulate algebraically my visual intuition about how to strip the turbulent eddies off of the backbone spatiotemporal attractors. One motivation is overcoming some of the deal-breaking limitations of EOF, CEOF, & VEOF analyses, which aren’t sufficiently generalized mode extractors. One inspiration is the work of Jose Rial (and possible extensions of that work). But as you can imagine, doing that sort of pioneering demands full-time, deep immersion. Working full time ruins it, but food and shelter take precedence over intellectual luxury!

    Cheers

  20. Erl, I have downloaded a pressure series for SE Qld from the NOAA site you suggest. Just looking through it without doing anything with the figures I observe
    1/ the data only goes back to 1948 and does not pick up most of the peak or extreme monthly rainfall events such as the 1890’s or the Federation drought 1901 to 1915 (particularly dry in 1905)
    2/ All the surface pressures appear to be on the high side in the range 1011.5 to 1019.9 mb (101.15 to 101.99 kPa) -atmospheric pressure at sea level should be 101.3 kPa. The pressures seem to be lower in the drier winter months July, August, September (ie in the range 1011.5 to 1013.5 mb and higher in the rainy summer months Jan, Feb, (ie 1014.5 to 1019.9mb which to me is the opposite of rain associated with low pressure . By contrast I have been looking at the SOI data which gives the Darwin and Tahiti surface pressures. Since the beginning of Jan 2016 (rainy season) the Darwin pressure has been in the range 1006.7 to 1011,8 mb. averaging around 1009mb.but Tahiti pressures around the same resulting in negative SOI (now moving towards zero showing the end of the El Nino)
    3/ Looking at the last 10 years where SE Qld has experienced a dry period (2001-2007) and above average rain (2009-2013) in including floods 2011(Brisbane) & 2012-2013 Gympie & Bundaberg there appears no relation between rainfall (i have measured) and the NOAA surface pressures
    I have have great doubts about the merit of the NOAA re-analysis data. -I could be rude and say crap data
    The SOI together with the PDO makes more sense.

  21. erl happ says:

    Cementafriend.
    Today, surface pressure across Queensland is 1014 to 1020mb according to this map: http://www.eldersweather.com.au/models/?lt=country&lc=aus&mt=gfs&mc=mslp&mh=0&focus=mh.

    Using the ESRL site I get 1013 to 1018mb for the April average. Looks pretty close to me.

    Bottom of the page should read something like this:
    Surface Sea Level Pressure (mb)
    Latitude Range used: -20.0 to -30.0
    Longitude Range used: 140.0 to 160.0

    Using the graphing function I get this: http://www.esrl.noaa.gov/psd/cgi-bin/data/timeseries/timeseries.pl?ntype=1&var=Sea+Level+Pressure&level=2000&lat1=-20&lat2=-30&lon1=140&lon2=160&iseas=0&mon1=0&mon2=0&iarea=1&typeout=2&Submit=Create+Timeseries

    Shows a jump in surface pressure after 1975 or thereabouts.

    Surface pressure has increased in August by 2mb since 1948. It increased in January and fell away to a no change situation for the period as a whole. Variability greater in August.I would expect it to be most variable in July and August, driven by change in the Arctic. But the long term drift is driven by the Antarctic.

    The ESRL series or other reanalysis series will be the only source for pressure data in remote ocean locations like the South East Pacific or the north Pacific.

    Darwin is probably a good proxy for the Indonesian area. But Tahiti is hopeless as a proxy for the south east Pacific.

    The Arctic Oscillation Index is a good proxy for Arctic surface pressure. But I don’t think it will relate well to the North Pacific.

    If you want to go back prior to the 1950’s you will have to rely on adjacent station data. eg Santiago Chile or Anchorage Alaska.These relate to the locations where surface pressure changes strongly affecting the planetary winds, ocean currents and therefore the amount of warm water in the tropics.

    Just a thought. Have you compared your rainfall data with the temperature of the waters off the coast or up in the Coral Sea.

    No, you haven’t shaken my faith in the validity of the reanalysis data.

  22. erl happ says:

    Cementafriend. I graphed surface pressure against precipitation rate. If atmospheric pressure falls in January you get rain. Very strong relationship.

    By and large in August its the other way round, you get rain when surface pressure rises.

    Now, I suggest you go look at surface pressure off the coast of Chile.

  23. ren says:

    April 19 jetstream will create high pressure north of the UK.

  24. ren says:

    Ionizing radiation shows the increase in pressure over Greenland and Iceland.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s