The Sun. Structure of Sun s Outer Layers Magnetism, Sunspots, Flares, and CMEs Solar Energy and Earth s Climate. 1 February 2018

Transcription

1 The Sun Structure of Sun s Outer Layers Magnetism, Sunspots, Flares, and CMEs Solar Energy and Earth s Climate 1 February 2018 University of Rochester

2 The Sun The structure of the Sun s outer layers: convection zone, photosphere, and corona Solar activity: magnetism, sunspots, flares, and CMEs Solar activity and Earth s climate Solar energy Reading: Ryden Ch. 14 (Atmospheres) and Ch. 15 (Interiors) Layers of the Sun (image from Hinode, NASA/JAXA). 1 February 2018 (UR) Astronomy / 1

3 The Solar convection zone The Sun s interior, being fully ionized, has γ = C P /C V = 5/3 Gas is unstable to convection if T dp P dt < γ 1 = 2 γ 5 In the Sun, this is true for 2 3 R < r < R, so it has a large outer convection layer. 1 February 2018 (UR) Astronomy / 1

4 The Solar convection zone Sunspot and solar granularity observed by Dutch Open Telescope (Rutten et al. 2017, APOD 2005). Each grain is the top of a convection cell. A 1-hour time lapse of the photosphere (van Der Voort 2016), see also here. 1 February 2018 (UR) Astronomy / 1

5 The solar dynamo & the solar cycle The Sun s radiative zone rotates like a solid body but the convection zone (CZ) rotates differentially. The differential rotation winds and amplifies a poloidal solar magnetic field, turning it into a more toroidal field (Higgins 2012). This occurs because the field is frozen into the ionized material by the Lorentz force. 1 February 2018 (UR) Astronomy / 1

6 The solar dynamo & the solar cycle Convection makes the field lines twist out of the surface and loop through the lower atmosphere, creating sunspot pairs and prominences connecting them. The twisting and winding of the field lines eventually causes the poloidal field to reappear but with the N and S poles reversed. 1 February 2018 (UR) Astronomy / 1

7 The solar dynamo & the solar cycle The reversal of the magnetic field repeats in a regular cycle. There are 22 years between identical configurations of the field. The self-generation process of the field is called dynamo action. For the Sun, there are 11 years between sunspot number maxima. From P. Charbonneu, U. Montreal (Charbonneau 2010). See also here. 1 February 2018 (UR) Astronomy / 1

8 The solar photosphere As we have noted, the spectrum of the sun closely resembles a blackbody. From the total energy flux at Earth total solar irradiance, (TSI), or solar constant f = erg s 1 cm 2 we get the Sun s luminosity L = erg s 1 Image from Mediawiki. TSI, L, and solar flux at most wavelengths vary little with time (which we will show). At very long and very short wavelengths, flares can change the Sun s brightness by huge factors. 1 February 2018 (UR) Astronomy / 1

9 The solar photosphere In detail: absorption lines are also seen in the solar spectrum. They match up with many known transitions of atoms, ions, and molecules. The solar spectrum. Nigel Sharp, using data from Bob Kurucz et al. (NOAO/NSO/Kitt Peak FTS/AURA/NSF). 1 February 2018 (UR) Astronomy / 1

10 The solar photosphere Spectral-line absorption by atoms and molecules is a hallmark of stars. Gases absorb strongly at the wavelengths of spectral lines (transitions between the quantum mechanical states) of the atoms and molecules of which they are composed. Stars are heated from inside and are cooler on the outside. Thus, to an outside observer, a star becomes opaque at a higher altitude for wavelengths of spectral lines. One sees deeper into the star at adjacent wavelengths. Because the deeper material is hotter, and hotter blackbodies are brighter, the star is brighter between the spectral lines. Hot Flux Cooler Wavelength 1 February 2018 (UR) Astronomy / 1

11 Chromosphere & corona The corona is heated by magneto-acoustic noise from the boiling top of the convection zone and by magnetic reconnection (flares). The coronal density is so low that it cannot cool efficiently, so it reaches temperatures > 10 6 K. From Chaisson and McMillan, Astronomy Today. 1 February 2018 (UR) Astronomy / 1

12 Corona during total solar eclipse Total solar eclipse, 2017, Carla Thomas, NASA. 1 February 2018 (UR) Astronomy / 1

13 Sunspots & solar activity Sunspots appear dark because they are slightly cooler than the rest of the solar surface. They are surrounded by hotter-than-average regions called faculae. Zeeman effect measurements show that they are maxima of the magnetic field. SOHO/NASA. 1 February 2018 (UR) Astronomy / 1

14 The 11-year sunspot cycle 1 February 2018 (UR) Astronomy / 1

15 Sunspots & solar activity Solar images from 29 Mar. 2001: He II at 30.4 nm Fe IX at 17.1 nm Visible light Note the X-ray bright faculae surrounding the sunspots. These regions add more to the Sun s luminosity than the spots take away. 1 February 2018 (UR) Astronomy / 1

16 Sunspots & solar activity Solar images from 29 Mar. 2001: He II at 30.4 nm Fe IX at 17.1 nm Visible light Note the X-ray bright faculae surrounding the sunspots. These regions add more to the Sun s luminosity than the spots take away. 1 February 2018 (UR) Astronomy / 1

17 Sunspots & solar activity Solar images from 29 Mar. 2001: He II at 30.4 nm Fe IX at 17.1 nm Visible light Note the X-ray bright faculae surrounding the sunspots. These regions add more to the Sun s luminosity than the spots take away. 1 February 2018 (UR) Astronomy / 1

18 Sunspots & solar activity X-ray emission from gas at T = 10 6 K in magnetic loops connecting the sunspots on the limb of the Sun (TRACE/NASA). 1 February 2018 (UR) Astronomy / 1

19 Sunspots & solar activity Flares are driven by magnetic reconnection within oppositely-opposed tubes of magnetic flux. The reconnected lines of B, strongly curved at first, straighten out quickly. The ions frozen to them impel other material outwards and inwards, like an arrow from a bow. Chen et al. (2008) 1 February 2018 (UR) Astronomy / 1

20 Sunspots & solar activity Large flares can lead to coronal mass ejection (CME) events, a.k.a. solar storms. CMEs are blasts of ions accelerated to high energy (v 0.3c) which expand into the Solar System. CMEs cause aurorae and wreak havoc on satellite and power systems on Earth. From STEREO/GSFC/NASA. 1 February 2018 (UR) Astronomy / 1

21 Notable CMEs September 1, 1859 The Carrington event. Strongest CME ever recorded. Flares visible to the naked eye; aurorae brighter than full moonlight and visible near the equator (Hawaii). Fires started in telegraph systems all over the world. August 2, 1972 Three strong CMEs in 15 hours, between the visits to the moon by Apollo 16 and 17. The astronauts would have been killed by the radiation had they been in space. November 4, 2003 Second-largest CME in history, one week after the fifth strongest in history, and a few months after the launch of the NASA Spitzer Space Telescope and two Mars Exploration Rovers (MERs) had taken these satellites outside the geomagnetic field. Damage to Spitzer s detectors: equivalent to 5 years of radiation damage in a few seconds. One MER s computer had to be rebooted 60 times before the memory repair routine finally worked. 1 February 2018 (UR) Astronomy / 1

22 Sunspots & solar activity The Sun s luminosity including flares and CMEs varies in sync with the sunspot number, currently by about 0.1% over the solar cycle. Data from Kopp et al. (2016) show irradiance data from about a dozen experiments: Irradiance in W/m 2 at top and sunspot number at bottom. Differences in irradiance are due to calibration differences between instruments. 1 February 2018 (UR) Astronomy / 1

23 The solar cycle & Earth s temperature The Earth s average equilibrium temperature can be estimated from P in = P out (1 A)f πr 2 = ɛσt 4 4πR 2 ( ) (1 1/4 ( A)f 1 A L T eq = = 4ɛσ ɛ 16πσr 2 Hence, a small change in solar luminosity L leads to a change T: T = dt L = 1 ( 1 A dl 4 ɛ = 1 ( 1 A L 4 ɛ 16πσr 2 ) 1/4 L 3/4 L 1 16πσr 2 ) 1/4 L 1 L = T L 4 L ) 1/4 1 February 2018 (UR) Astronomy / 1

24 The solar cycle & Earth s temperature The 150-year average of the global mean ocean surface temperature (GMOST) is 16.0 C = 289 K, and the ocean s heat capacity is such that it can respond to heating changes in about a month. That is why it is usually colder at the end of January than on December 21. A month is short compared to 11 years, so we expect T = K = K = C per 1 W m 2 change in TSI, or equivalently per change by 100 in annual sunspot number. 1 February 2018 (UR) Astronomy / 1

25 Sunspots and climate on multi-solar cycle scales Sunspots have been counted systematically for hundreds of years. Over that span there have been big swings in the peak sunspot number in solar cycles. There has been one extended period centered around the 17 th century called the Maunder Minimum, when sunspots were practically absent. Over that span there have also been big swings in Earth s climate. One extended period occurred during the 17 th century. Known as the Little Ice Age, the Earth s surface was dramatically colder than today. These coincidences have not gone unnoticed. 1 February 2018 (UR) Astronomy / 1

26 Annual sunspot number for the last 360 years Annual sunspot number200 Little Ice Age Maunder Minimum Historically harsh winters (novels by Hugo, Dickens,...) Year Sources: NOAA/NCEI and SIDC/SILSO. 1 February 2018 (UR) Astronomy / 1

27 Earth s global surface temperature To match up temperature with sunspots and their record of total solar irradiance, we have several ways of monitoring global mean ocean surface temperature, GMOST. We have global ocean surface temperature measurements taken with real thermometers going back to We have satellite measurements since the 1960s. We also have ways to recover oceanic temperatures over much longer time spans: Water-isotope abundance measurements in ice cores and tree rings. Tree ring width (e.g., Mann & Jones 2003). All of these techniques agree with the historical record for the last two millenia. 1 February 2018 (UR) Astronomy / 1

28 The last 1800 years of T and sunspots N. Hemisphere T 16 C Climatic optimum (Vikings settle Greenland) Little Ice Age year annual sunspot number Sources: Mann & Jones (2003), NASA/GISS, NOAA/NCEI, and SIDC/SILSO. 1 February 2018 (UR) Astronomy / 1

29 Sunspots & climate For much of the last 400 years, the temperature and peak sunspot number track with each other: eras with high (low) peak sunspot number go with eras with high (low) temperature. You might think this suggests that variation in solar activity produces variation in climate. There are at least three big problems with that suggestion: 1. Ice ages are even colder ( 8 C) than Little Ice Ages ( 0.4 C). Extrapolating you cannot get negative sunspot numbers. 2. The correlation deteriorates continuously through the past century; it vanished a few decades ago. 3. Some of the more recent temperature drops are much better explained by volcanic activity: absorption of sunlight by high-altitude ash and sulfuric acid (H 2 SO 4 ). 1 February 2018 (UR) Astronomy / 1

30 The last 1800 years of T and volcanic explosions N. Hemisphere T 16 C Taupo Ilopango Baekdu year Sources: Mann & Jones (2003), NASA/GISS, and MediaWiki. Kuwae Kolumbo Mt. Tambora volume of tephra [km 3 ] 1 February 2018 (UR) Astronomy / 1

31 T, solar cycle, & volcanic explosions, N. Hemisphere T 16 C Kolumbo year Sources: Mann & Jones (2003), NASA/GISS, and MediaWiki. Mt. Tambora Krakatoa volume of tephra [km 3 ] 1 February 2018 (UR) Astronomy / 1

32 T, solar cycle, & volcanic explosions, 1900 Present N. Hemisphere T 16 C Pinatubo year volume of tephra [km 3 ] Sources: Mann & Jones (2003), NASA/GISS, and MediaWiki. 1 February 2018 (UR) Astronomy / 1

33 Sunspots & climate Solar activity does not dominate the long-term substantial changes in global temperature, neither global warming nor cooling periods. Better explanations: Increases in atmospheric CO 2 since the start of the Industrial Revolution (1750s) almost certainly have caused the recent trend of average global warming. Large volcanic explosions in the tropics seem at least as promising an explanation of the Little Ice Age as the low solar activity during the Maunder Minimum. Volcanic explosions also appear to explain some of the more recent and confidently characterized (minor) global cooling events. 1 February 2018 (UR) Astronomy / 1

34 The last 1800 years of T & atmospheric CO 2 N. Hemisphere T 16 C year Sources: Mann & Jones (2003), NASA/GISS, Etheridge et al. (1998), and NOAA ESRL/Scripps atmospheric CO2 [ppm] 1 February 2018 (UR) Astronomy / 1

35 T & atmospheric CO 2, 1800 Present N. Hemisphere T 16 C year atmospheric CO2 [ppm] Sources: Mann & Jones (2003), NASA/GISS, Etheridge et al. (1998), and NOAA ESRL/Scripps. 1 February 2018 (UR) Astronomy / 1

36 Solar activity & climate on solar-cycle timescales What about smaller temperature changes on short timescales? Sunspot number, TSI, and temperature are also correlated on timescales shorter than the solar cycle. This results in a temperature change of 0.15 C per 1 W m 2 change in TSI over the 160-year thermometer record annual sunspot number year annual TSI [W m 2 ] Results on the 35-year TSI record yield a similar result: 0.11 C per W m 2 of TSI. Either way, it is significantly more than the solid expectation of 0.05 C calculated several minutes ago. 1 February 2018 (UR) Astronomy / 1

37 Solar activity & climate on solar-cycle timescales Source: NOAA/NCEI 1 February 2018 (UR) Astronomy / 1

38 Solar activity & climate on solar-cycle timescales 1 February 2018 (UR) Astronomy / 1

39 Solar activity & climate on solar-cycle timescales The amplification of TSI changes is puzzling; it cannot be accounted for by the current generation of climate models. It would be a worry even without global warming! The observed T in phase with the solar cycle cannot be produced without a factor of 5 to 7 more heat than the change in TSI. Climate models fail many simple tests like this. Like the climate itself, they are extremely complex and just do not yet work perfectly. The models will get better, but should you be worried? Yes. We cannot really predict the size of anthropogenic effects if we cannot correctly predict the size of temperature modulation from the tiny solar-cycle modulation. 1 February 2018 (UR) Astronomy / 1

40 A recent model disentangling contributions to global temperatures Right: natural and manmade influences on surface temperature, from Jean & Rind (2008). ENSO = El Niño Southern Oscillation. The peaks in 1983, 1997, and 2015 were very strong El Niño events. Volcanic aerosols spike and cool the global climate, but then decay. Effects by solar irradiance are cyclical and have been at the ±0.1 C level recently. The secular trend upwards is tracing steady increases in anthropogenic greenhouse gases (mostly CO 2 ). 1 February 2018 (UR) Astronomy / 1

41 Energy & the Sun A thought experiment: Hydrostatic equilibrium and ideal gas behavior ensure that the center of the Sun is very hot, and energy (in the form of light) is radiated from the center. The high opacity of the Sun to light determines the rate at which energy leaks out. As we have seen, it takes a long time for photons to diffuse from the center to the surface. This cannot go on forever without the Sun cooling down or replacing the energy that leaks away. We know that the solar system is about years old from many radioisotope measurements of meteorites, and that life has existed here for at least years. Therefore the Sun must have had close to its present luminosity for billions of years for liquid water to be present on Earth to facilitate life. 1 February 2018 (UR) Astronomy / 1

42 How long would the Sun s present heat last? The energy density at the center of the Sun is given by the energy density of the electron gas there, considered to be an ideal gas: u e = 3 2 n ekt The energy density of the radiation (light) about to leak away is u r = 4 c f = 4σT4 c We showed last time that it takes t years for a photon to leak from the center to the surface, so the heat lasts u e du e /dt u e u r /t = 3 2 n ekt c 4σT 4 t 3 8 kc ρ σm p T 3 t yr This is much less than the Sun s age so some process must be replacing the energy that leaks away. 1 February 2018 (UR) Astronomy / 1