From the UK Telegraph – source link
The protective bubble around the sun that helps to shield the Earth from harmful interstellar radiation is shrinking and getting weaker, NASA scientists have warned.
By Richard Gray, Science Correspondent
Last Updated: 9:23AM BST 19 Oct 2008
New data has revealed that the heliosphere, the protective shield of energy that surrounds our solar system, has weakened by 25 per cent over the past decade and is now at it lowest level since the space race began 50 years ago.
Scientists are baffled at what could be causing the barrier to shrink in this way and are to launch mission to study the heliosphere.
The Interstellar Boundary Explorer, or IBEX, will be launched from an aircraft on Sunday on a Pegasus rocket into an orbit 150,000 miles above the Earth where it will “listen” for the shock wave that forms as our solar system meets the interstellar radiation.
Dr Nathan Schwadron, co-investigator on the IBEX mission at Boston University, said: “The interstellar medium, which is part of the galaxy as a whole, is actually quite a harsh environment. There is a very high energy galactic radiation that is dangerous to living things.
“Around 90 per cent of the galactic cosmic radiation is deflected by our heliosphere, so the boundary protects us from this harsh galactic environment.”
The heliosphere is created by the solar wind, a combination of electrically charged particles and magnetic fields that emanate a more than a million miles an hour from the sun, meet the intergalactic gas that fills the gaps in space between solar systems.
At the boundary where they meet a shock wave is formed that deflects interstellar radiation around the solar system as it travels through the galaxy.
The scientists hope the IBEX mission will allow them to gain a better understanding of what happens at this boundary and help them predict what protection it will offer in the future.
Without the heliosphere the harmful intergalactic cosmic radiation would make life on Earth almost impossible by destroying DNA and making the climate uninhabitable.
Measurements made by the Ulysses deep space probe, which was launched in 1990 to orbit the sun, have shown that the pressure created inside the heliosphere by the solar wind has been decreasing.
Dr David McComas, principal investigator on the IBEX mission, said: “It is a fascinating interaction that our sun has with the galaxy surrounding us. This million mile an hour wind inflates this protective bubble that keeps us safe from intergalactic cosmic rays.
“With less pressure on the inside, the interaction at the boundaries becomes weaker and the heliosphere as a whole gets smaller.”
If the heliosphere continues to weaken, scientists fear that the amount of cosmic radiation reaching the inner parts of our solar system, including Earth, will increase.
This could result in growing levels of disruption to electrical equipment, damage satellites and potentially even harm life on Earth.
But Dr McComas added that it was still unclear exactly what would happen if the heliosphere continued to weaken or what even what the timescale for changes in the heliosphere are.
He said: “There is no imminent danger, but it is hard to know what the future holds. Certainly if the solar wind pressure was to continue to go down and the heliosphere were to almost evaporate then we would be in this sea of galactic cosmic rays. That could have some large effects.
“It is likely that there are natural variations in solar wind pressure and over time it will either stabilise or start going back up.”
(hat tip to Dvid Gladstone)
egrey (22:06:15) :
GCRs/clouds/[…] The latter has not been eliminated.
GCRs were supposed to control the low clouds. Here is observational refutation of that idea:
http://www.climate4you.com/images/CloudCoverAllLevel%20AndWaterColumnSince1983.gif
The variation does not follow the [inverse] solar cycle/GCR flux.
Leif
GCRs were supposed to control the low clouds
You can not compare pre- to post-2001 directly (2002 in my prev post is wrong, it should be oct 2001)
There are plenty of evidence and ISCCP explain it on their pages.
Here’s a couple of examples: http://virakkraft.com/NET-SRF-low-unadj.jpg
Clearly the jumps late 2001 are not real.
Nor can you expect to see the sun-cloud signal in the albedo, which is the total of several signals, volcanism in particular. The eruptions early 80s and 90s preclude the signal, in addition, the lack of large eruptions second half of the 90s makes it even worse.
lgl (11:24:19) :
Clearly the jumps late 2001 are not real.
Is the definition of ‘not real’ that it doesn’t fit the theory?
Nor can you expect to see the sun-cloud signal in the albedo
I thought that was the mechanism, no?
There is a clear relationship between cloud cover and albedo:
http://www.leif.org/research/albedo.png
http://www.leif.org/research/cloud-cover.png
Plot of low clouds show no solar cycle [GCR] signal:
http://www.climate4you.com/images/CloudCoverAllLevel%20AndWaterColumnSince1983.gif
lgl (11:24:19) :
There are plenty of evidence and ISCCP explain it on their pages.
Their plot of low clouds:
http://isccp.giss.nasa.gov/zD2CLOUDTYPES/B32glbp.anomdevs.jpg
The data up to ~2000 shows the correlation that Svensmark latched on to. If the correlation was due to the underlying cause of solar cycle modulation of GCRs, then it should continue in time after its first discovery. As is so often the case with Sun-Climate correlations, it didn’t. This is then observational refutation.
As Donald R. said: “There are known knowns; there are things we know we know. We also know there are known unknowns”.
Leif
Is the definition of ‘not real’ that it doesn’t fit the theory?
No it is when it does not fit anything. Take a look here then: http://isccp.giss.nasa.gov/projects/browse_fc.html
NET-SRF for instance, late 2001, very likely?
NET-TOA, a cooling after 2002 of Pinatubo magnitude, very likely?
I contacted ISCCP and they confirmed the drop late 2001 is not real, you have to subtract the similar step-up of LW UP TOA late 2001, which is not real either. In NET-TOA you see the impact of volcanoes, also found in ALBEDO.
I thought that was the mechanism, no?
I thought the mechanism was modulation of low level cloud cover, albedo is much more.
http://isccp.giss.nasa.gov/zD2CLOUDTYPES/B32glbp.anomdevs.jpg
The drop late 2001 is probably mainly caused by the lat. -30,-60 drop I have shown, it’s not real.
lgl (14:03:18) :
The drop late 2001 is probably mainly caused by the lat. -30,-60 drop I have shown, it’s not real.
Nowhere on the ISCCP website do I see any mention of this. One would think that such a serious error would be front and center, no?
The website has a list of known errors:
1. The BT for METEOSAT-7 from 06/1998 to 09/2001 for IR channel has a couple of systematic errors.(have been corrected)
2. The clear sky composite for 1.6 micron channel is biased a little low.
3. Missing DX Geostationary 1st additional channel [ARAD(1)]
4. Spurious satellite zenith angle dependence (artifact in Indian sector)
5. Change in land surface pressures from TOVS
6. Incorrect snow cover (has been corrected)
7. Incorrect precipitable water amounts and surface temperatures (has been corrected)
8. Spurious sea ice reports
9. Flipped array indices in North Polar DX (has been corrected)
10. Spurious land pixels in METEOSAT DX
11. D2 versions 0,1,2 contain un-corrected METEOSAT-3 reflectances (has been corrected)
12. Cloud top temperature/pressure error
13. Change to daily atmospheric temperature profiles
14. Error in monthly-mean atmospheric temperature profiles
15. Missing GOES-7 visible data
16. Error in normalized calibration of GOES-8/9 water vapor channel
17. Systematic decrease in surface temperatures due to changes in NOAA operational sounder analysis
Which one(s) of these errors is the ‘step’ related to?
Once we are down to discussing errors in the various datasets that are available to the public [via government websites], I usually take a step back and look at my understanding of how things should work. For a long time, my main argument against the Svensmark idea has been this: The Earth’s magnetic field has decreased 10% in the last 150 years. That means that GCRs should have increased over that time with a steady cooling as the result. Such cooling over the last 150 years has not been observed, most people would claim instead a significant warming. This is independent of dataset, step correction, volcanoes, albedo, etc.
To me this is so devastating for the GCR theory that one can stop right there.
An amusing example of the contortions people go through trying to salvage a dead relationship can be found here:
http://lasp.colorado.edu/sorce/news/2005ScienceMeeting/presentations/thur_am/Stager_Lake_Victoria.pdf
From 1896 to 1927 AD, a strong (0.87) correlation existed between SSN and the levels of Lake Victoria, then the correlation went away. This is the classic textbook example of the ‘case of the disappearing correlation’. Much later it seemed that the correlation surfaced again [and became ‘visually appealing”. Page 15 of the powerpoint presentation is particularly telling. Note how the correlation is back! But also note the strange red curve [the SSN], hint: look when minima and maxima are for this ‘SSN’ curve!
Leif – thanks for the various posts. It’s going to take me a while to work through the information. I appreciate the time you have spent on the issues raised here, and the extra food for thought that you have supplied.
Leif Svalgaard (15:31:28) :
The Earth’s magnetic field has decreased 10% in the last 150 years. That means that GCRs should have increased over that time with a steady cooling as the result. Such cooling over the last 150 years has not been observed, most people would claim instead a significant warming. This is independent of dataset, step correction, volcanoes, albedo, etc.
To me this is so devastating for the GCR theory that one can stop right there.
Even looking at your reconstruction of 10Be records and aa records shows a rise in solar magnetism up to around 2000 that would surely outweigh the reduced earth’s magnetic field?
It would be interesting to see a combined graph.
nobwainer (18:30:56) :
To me this is so devastating for the GCR theory that one can stop right there.
Even looking at your reconstruction of 10Be records and aa records shows a rise in solar magnetism up to around 2000 that would surely outweigh the reduced earth’s magnetic field?
We can do better than that. We have values of the Sun’s magnetic field [in the Heliosphere where it counts] since 1836. See page 20 of http://www.leif.org/research/Seminar-SPRG-2008.pdf and read the whole paper if you want to know how we derive the heliospheric field, HMF.
It is clear that the HMF the last 30 years is no different from what it was in the mid-1900s. Cycle 23 is almost a replica of cycle 13, 107 years ago [page 26]. So there has been no steady upwards change of the HMF.
It would be interesting to see a combined graph.
Leif Svalgaard (18:51:01) :
nobwainer (18:30:56)
The last para should have been:
“It is clear that the HMF the last 30 years is no different from what it was in the mid-1800s. Cycle 23 is almost a replica of cycle 13, 107 years ago [page 26]. So there has been no steady upwards change of the HMF.”
Not 1900s; I was thinking of 19th century, of course. Grrr. Wouldn’t it be handy with a preview/edit function…
Interesting paper Leif…not sure i understand it all but get your drift. Your method of reconstruction if proven accurate would certainly change how we calculate any solar involvement in GCR theory….but what i cant quite understand is if the earth’s magnetic field is down 10% and the sun has remained fairly static why are we seeing a steady decrease in the 10Be records that coincides with sunspot activity over the past century or so?
One explanation could be the GCR’s are themselves fluctuating as we circle the galaxy…but the 10Be records do seem to line up nicely with the sunspot records?
nobwainer (20:44:02) :
why are we seeing a steady decrease in the 10Be records that coincides with sunspot activity over the past century or so?
We argue that the 10Be records are not calibrated correctly:
http://www.leif.org/research/Comment%20on%20McCracken.pdf
http://www.leif.org/research/TSI%20From%20McCracken%20HMF.pdf
nobwainer (20:44:02) On about a 60 million year cycle our solar system moves in and out of the band of maximum galactic radiation. I think we’re about 10 million years now from a maximum.
=========================================
kim (07:19:50) A tip of the toppah to Adrian Melott a for thattah.
=======================================
Leif
Which one(s) of these errors is the ’step’ related to?
Regarding radiation, I guess this: http://isccp.giss.nasa.gov/projects/browse_fc.html
“However, the sudden increase in upwelling LW flux in late 2001 may be exaggerated because it is associated with a spurious change of the atmospheric temperatures in the NOAA operational TOVS products that are used in the calculations.
I don’t know how the radiation data relates to the cloud data but when both sets start jumping all over the place at the same time even my simple brain is able to link the two.
Here’s one from -30,-60 high clouds: http://virakkraft.com/high-clouds.jpg
high clouds always peak mid-year and decrease towards end-of-year, but not in 2001. peak middle of year, start to decrease then suddenly increase 2 % and then continue the normal decrease until end of year. Errors just don’t get more visible than this. Believe whatever you need to feel comfortable but the sun-cloud link is not refuted by observation.
Leif,
… one more thing, The sun is not the only player. If you expect to find a perfect correlation between cloud cover and solar activity you assume that all other factors have remained constant over the same period, at least volcanism has not.
lgl (10:40:49) :
“However, the sudden increase in upwelling LW flux in late 2001 may be exaggerated because it is associated with a spurious change of the atmospheric temperatures in the NOAA operational TOVS products that are used in the calculations.
I don’t know how the radiation data relates to the cloud data but when both sets start jumping all over the place at the same time even my simple brain is able to link the two.
First, they don’t say that the step is spurious, just that it is bigger than it should be, so there is a real step too. Second, the cloud analysis does not depend on the temperatures, so although your ‘simple brain’ sees what it wants, that may be jumping to conclusions.
Here’s one from -30,-60 high clouds: […] Errors just don’t get more visible than this.
If you think 2001 has errors, omit the year. If you do that, the variation doesn’t look too bad. Now, when Svensmark and company make there correlations do they also use ISCCP? If so, isn’t their correlation marred by the same ‘errors’?
The sun is not the only player. If you expect to find a perfect correlation between cloud cover and solar activity you assume that all other factors have remained constant over the same period, at least volcanism has not.
I most certainly don’t expect to find such a correlation. I do expect to find a good correlation between cloud cover and albedo and there is such a good correlation. The albedo does not correlate with solar activity, so now we have to say that other factors [volcanoes] determine the albedo to the extent that the solar connection is lost in the noise.
You did not react to my main point, so I’ll repeat it here and hope for a reaction next time:
Once we are down to discussing errors in the various datasets that are available to the public [via government websites], I usually take a step back and look at my understanding of how things should work. For a long time, my main argument against the Svensmark idea has been this: The Earth’s magnetic field has decreased 10% in the last 150 years. That means that GCRs should have increased over that time with a steady cooling as the result. Such cooling over the last 150 years has not been observed, most people would claim instead a significant warming. This is independent of dataset, step correction, volcanoes, albedo, etc.
To me this is so devastating for the GCR theory that one can stop right there.
Leif
I know you too see that 2001 looks totally different all other years. No other year shows an icreased high level cloud cover second half of the year (and it appears to happen around october but I haven’t checked the numbers)
I don’t have any opinion on the GCR theory, it’s too new and unproven and I don’t have the knowledge. And I have no idea how reliable 150 years old records of earth’s magnetic field are, nor do I know what impact it will have at different latitudes. But I know the oct. 2001 ‘event’ makes it impossible to compare before and after directly and that makes the data useless for refuting the sun-cloud connection, and it’s not necessarily a GCR-cloud connection.
lgl (13:57:00) :
But I know the oct. 2001 ‘event’ makes it impossible to compare before and after directly and that makes the data useless for refuting the sun-cloud connection, and it’s not necessarily a GCR-cloud connection.
makes THAT data useless for refutation, but also for claiming the connection. But, there are other data and none I have seen support the GCR-connection. [I think it was you that asked me to look at the ISCCP data].
and it’s not necessarily a GCR-cloud connection.
Tell Svensmark that 🙂
This was billed as a clear-cut GCR-cloud connection from the beginning IIRC, then it became low-clouds only, and now it is not necessarily a GCR-cloud connection…
And I have no idea how reliable 150 years old records of earth’s magnetic field are, nor do I know what impact it will have at different latitudes.
But I do, and this is textbook stuff.
Leif…I read your McCraken papers, but did not find your arguments compelling. Your work on reconstructing the 10Be, geomagnetic aa, sunspot counts and TSI is in my opinion controversial and needs to be supported more by further research….but wish you luck.
Basically from what i have read you are suggesting a floor for the solar proxy records arguing that solar activity is predominately flat. I have difficultly with your outcomes and in particular wonder how your theory stacks up against the temperature record from 1900 on. If solar activity is not linked to temperature then what accounts for the global rise from 1900 to 1950 and then the decline towards 1970 and then the rise in the 80’s and 90’s (that curve correlates nicely to the other 10Be, aa, TSI and sunspot records) I know there are other factors (volcanoes, nuclear testing, sulfur dioxide and PDO, AMO, ENSO etc) but i would not see them as the driver of climate but more as secondary mechanisms.
My observations lead me to think that both the Sun and Earth are not totally autonomous and ARE affected by outside influences…this is probably in direct contrast to your observations but look forward to the puzzle being solved.
nobwainer (16:45:52) :
I read your McCraken papers, but did not find your arguments compelling.
Let’s take one thing at a time. The very short Comment on McCracken has three sections each setting out one argument. What are your specific problems with those three? There is a simple rule in peer-reviewing that it is not enough to say that an argument is not compelling, or wrong, or weak, or whatever. The reviewer must also state why the argument is not compelling, wrong, weak, or whatever. Only in that way can the author respond in any meaningful manner.
Something that was not very clear in the version on my website is that Rouillard and Lockwood agree [personal communication] that their 1901 data point is in error. I have uploaded the revised version, also taking into account comments from the referee.
Leif Svalgaard (17:34:43) :
The reviewer must also state why the argument is not compelling, wrong, weak, or whatever.
————————————————————-
1. The Lockwood et al. [1999] reconstruction has been superceded, resolving the disagreement with Svalgaard and Cliver [2005]
While encouraging its one paper that is closer to yours and if i am not mistaken compares aa records…it wasnt clear to me what brought about their reconstruction? I have seen many instances where the aa data and 10Be records dont align…around 1830 and 1840 in particular.
2. The Solanki et al. reconstruction is not independent of Lockwood et al.
If it is proven that Solanki tweaked the model to agree with Lockwood’s earlier work then it would be one area McCracken would lose support. Are there other papers that support McCracken and what other 10Be studies have been done that might support McCracken? this is not clear in your comment.
3. The McCracken 1426-2005 HMF reconstruction needs to be re-examined
You cast doubt on the 1933-1951 portion of McCrackens data but mainly disagree because it does not match your findings and Rouillard et al. This is much like point 1. If this portion is found to be incorrect would that also negate the 1428-1930 section that also shows many dips and would that section need to be adjusted up in amplitude or would it remain?
You have perhaps showed some weaknesses in 2 papers that supported McCrackens paper but that does necessarily mean his paper is in error.
kim (07:19:50) :
nobwainer (20:44:02) On about a 60 million year cycle our solar system moves in and out of the band of maximum galactic radiation. I think we’re about 10 million years now from a maximum.
Our galaxy orbit time is estimated at around 226 millions years which suggests multiple bands of high galactic radiation…..glad we have 10 million years up our sleeve altho i thought i read somewhere recently that we are due for an iceage based on previous cycles…might have to research that further.