Modeling and Measurement of Upper Atmosphere Climate Change

White Paper submitted to the Decadal Survey of Solar and Space Physics

Stanley C. Solomon and Liying Qian High Altitude Observatory National Center for Atmospheric Research

Summary Human activity has caused subtle changes in the Earth’s atmosphere over the past several decades, due to increased emissions of gases such as carbon dioxide (CO2) and methane (CH4). These anthropogenic changes in greenhouse gases have the potential for increasing temperature in the lower atmosphere due to their ability to absorb infrared radiation. The far-reaching conse- quences of these changes are now being seen in the upper atmosphere of the Earth, especially in the thermosphere. Since the prediction of Roble and Dickinson [1989] that a consequence of increasing CO2 levels would be to decrease the temperature of the upper atmosphere, opposite to the response of the lower atmosphere, several investigators have attempted to quantify observed effects on the mesosphere, thermosphere, and ionosphere. In the thermosphere, one effect of a cooler atmosphere is to cause global density to decrease as the thermosphere cools and contracts. This in turn causes changes in the ionosphere, lowering ionospheric layer heights and changing their density. The thermosphere goes through natural, cyclical changes in density driven by the Sun’s 11-year activity cycle, heating and expanding at solar maximum, cooling and contracting at solar minimum, which makes modeling of secular change a challenging process, due to the interaction of solar and anthropogenic effects. In addition, the evolving terrestrial magnetic field influences ionospheric change, and the role of magnetosphere-ionosphere coupling must be con- sidered. In order to quantify and understand global change in the upper atmosphere, three- dimensional general circulation models of the thermosphere-ionosphere-mesosphere system should be integrated with global Earth system models that include the entire atmosphere, and coupled to interactive ocean and land mass models. To quantify the fundamental physics of the upper atmosphere radiative energy balance, space-based measurements of infrared emissions and cooling rates must be performed, and spectrally-resolved measurements of the full-disk solar irradiance at ultraviolet and X-ray wavelengths need to continue. Observational data of middle atmosphere, thermosphere, and ionosphere change on solar-rotational, solar-cycle, and secular trend time scales should be analyzed and used to improve model fidelity and quantify uncertain aeronomical processes. Objectives Global change in the upper atmosphere is an important issue for at least three basic reasons: it is important to understand the full physical scope of anthropogenic change in the entire atmosphere, in order to establish thorough understanding of lower atmosphere effects; there are practi- cal implications for low-Earth orbit space flight, satellite drag, and the accumulation of orbiting debris; the interplay of external forcing and the natural variability of the atmosphere with changes caused by anthropogenic trace gas emissions is an inherently compelling scientific topic for heliophysics. In order to fully understand and quantify secular change of thermospheric temperature and density, we must also obtain a complete and accurate description of solar and geomagnetic forcing. An integrated program using theoretical models, solar measurements, satellite drag derived inferences of thermospheric neutral density, ionospheric measurements, and infrared observations of thermospheric cooling rates, will be able to address key questions concerning global change in the upper atmosphere:

1 • Is the secular change in thermospheric density observed over the past ~40 years due only to anthropogenic changes in atmospheric constituents, or are changes in solar and geomagnetic activity also implicated? • How does solar variability, the amplitude of the solar cycle, and the length and depth of solar minimum, interact with upper atmosphere climate change? • How will anthropogenic change in the ionosphere be manifested in layer heights and densities, E-region conductivities, F-region electrodynamics, and ion and electron temperatures? • How will changes in the terrestrial magnetic field influence magnetosphere-ionosphere coupling, auroral dynamics, ionospheric morphology, and geospace climate? • Do variabilities in the lower- and middle-atmosphere, such as the QBO, the El Niño south- ern oscillation, and stratwarms, modulate mesosphere-thermosphere-ionosphere climate? • Will hydrogen in the exosphere increase as a consequence of global change, and what would be the effect on the plasmasphere and magnetosphere? Background The temperature and density of the upper atmosphere are determined by the interaction of energetics, dynamics, and composition. The absorption of solar radiation is the primary heat source, and as the wavelength of solar photons decreases, they are generally absorbed at higher altitude. Solar ultraviolet irradiance varies on periods associated with the 11-year solar cycle and the 27-day solar rotational period with the variation increasing with decreasing wavelength. Thus, the atmospheric effects of solar variability increase with altitude, with higher temperature causing higher density at a constant altitude surface, due to the expansion of the atmosphere. Solar EUV is completely absorbed in the thermosphere due to the large ionization cross- sections of the major species. The products of ionization and dissociation initiate exothermic chemical reactions that determine time-varying neutral composition and ion composition. These photoionization and photodissociation processes thus transfer photon energy to chemical potential energy and kinetic energy that heats the neutral atmosphere, creates the ionosphere, and dis- sociates the major molecular atmospheric gases, N2 and O2, into atoms. Orbits of satellites in the thermosphere change due to the frictional drag generated by collisions with these atoms and molecules; the amount of orbital decay is determined primarily by the neutral density. Thus, satellite orbit decay caused by atmospheric drag is both a technical problem we wish to address through the use of prognostic modeling techniques, and, through analysis of change of orbital elements, a means by which to measure the thermosphere and thereby understand its response to external forcing and internal processes. Additional variability is caused by magnetospheric energy input at high latitude, the propagation of waves and tides from the lower/middle atmosphere, and the resulting neutral winds. Thermospheric neutral density routinely has daily changes normally on the order of 10%, but sometimes as large as a factor of two, and can have a solar cycle variation of an order of magni- tude at altitudes in the 400–600 km range. Empirical models describe this variation by character- izing the temperature and density changes as a function of indexed parameters, and first- principles theoretical models, including general circulation models, have attempted to establish a theoretical description through quantification of the energy sources and sinks and the chemical and dynamical processes through which they interact [e.g., Dickinson et al., 1981; Fuller-Rowell et al., 1987; Roble et al., 1988; Richmond et al., 1992; Roble and Ridley, 1994]. Superimposed on this solar-driven variation is a gradual decrease in temperature and density caused by increasing CO2 and other anthropogenic “greenhouse gases” (GHG). Roble and Dick- inson [1989] first predicted that a consequence of increasing CO2 levels would be to decrease the temperature of the upper atmosphere, opposite to the response of the lower atmosphere. The rea- son for this apparent paradox is that CO2 and other heteronuclear molecules can emit infrared

2 radiation as well as absorb it. In the lower atmosphere, CO2 absorbs radiation coming from the Earth, which excites it to higher vibrational states. Before it can re-emit the radiation, it under- goes collisions with other atmospheric gases, transferring the vibrational energy into heat. Any emission that does occur is likely to be re-absorbed by CO2 at other altitudes. But above the tro- popause the atmosphere becomes increasingly transparent to infrared radiation as densities decrease, and the radiation-excitation-collision process runs in reverse. Instead of absorbing radiation, CO2 molecules are vibrationally excited by collisions, and then spontaneously emit in the infrared, which causes radiational cooling of the upper atmosphere.

CO2 is the most important, but not the only, GHG with the potential to affect the upper atmosphere. CH4 is both radiationally active in the infrared and a source of hydrogen to the mesosphere and thermosphere, which may affect polar mesospheric cloud (noctilucent cloud) frequency, odd-hydrogen in the mesosphere, atomic hydrogen in the exosphere, and hydrogen escape. Chlorine and nitrogen oxides in the stratosphere change the O3 balance and affect both odd-oxygen chemistry and stratospheric heating. Changes in water vapor consequent to tropo- spheric temperature change can also propagate through the middle atmosphere. Thus, other potential anthropogenic effects include mesosphere cooling, increases in the frequency and latitu- dinal extent of polar mesospheric clouds, increasing atomic hydrogen in the exosphere, and ionospheric changes. Recent reviews describe these and other possible effects, and assess the observational evidence [Beig et al., 2003; Lastovicka et al., 2006; 2008]. The thermospheric density trend predicted by Roble and Dickinson [1989] has been observed in long-term studies of satellite orbits [Keating et al., 2000; Emmert et al., 2004; Marcos et al., 2005; Lean et al., 2006; Emmert et al., 2008]. These studies have unambiguously observed these trends, with the changes in upper thermospheric density ranging from ~2% to ~5% per decade, depending on altitude and solar activity. Other model studies [e.g., Akmaev and Fomichev, 1998, 2000; Akmaev et al., 2006; Gruzdev and Brasseur, 2005] similarly predicted a downward secular trend in thermosphere and mesosphere temperatures. Qian et al. [2006] explained the solar cycle dependence as due to modulation by nitric oxide cooling rates, using an updated version of the Roble et al. [1987] global mean model, and obtained good agreement with some of the observational data. However, the latest analyses [Emmert et al., 2008] confirms the large changes for solar minimum conditions, 5% per decade or more, first reported by Keating et al. Thus, the observational evidence, while confirming thermospheric cooling, indicates that models may under- estimate the rate of change. Figure 1 compares observational and model estimates of the rate of thermospheric density change as a function of solar activity.

Figure 1. Left: Global average thermospheric density residuals at 400 km obtained from atmospheric drag on ~5000 orbiting objects, and inferred secular trend [Emmert et al., 2008]. Right: Measurement and model estimates of thermospheric density trends as a function of solar activity [Lastovicka et al., 2008].

Additional evidence comes from radar measurements of ion temperature. Holt and Zhang [2008] found a trend of -47 K per decade in ion temperatures at 350–400 km measured by the incoherent scatter radar at Millstone Hill, MA. This implies a neutral temperature decrease significantly larger than expected from model simulations or seen in satellite drag measurements, and, if substantiated, would have significant implications for the ionosphere as well. Changes in the ionosphere should accompany the neutral atmosphere changes, including changes in the height of ionospheric layers due to cooling and contraction, and small changes in peak density [Rishbeth and Roble, 1992]. Since ionospheric measurements, especially ionoson- des, have a long history, these records could provide a means of assessing secular change. Peak densities at E and F region altitudes (NmE and NmF2) are easier to measure in a stable long-term manner than the peak altitudes (hmE and hmF2), since they are based on critical frequency. How- ever, model predictions of peak density change reveal a complex scenario, due to the interaction of atmospheric contraction with changes in composition and chemical rates. The E and F1 re- gions are expected to actually increase slightly in density as they reduce in altitude, and very small changes are anticipated in NmF2. Observational evidence is reviewed by Lastovicka et al. [2008]; model results and the underlying theory are described by Qian et al. [2008; 2009]. Ionospheric secular change is also complicated by evolution of the terrestrial magnetic field. Over the past century, the dipole strength has decreased about 6%, and non-dipole features of the field also evolve, affecting magnetic declination and inclination angles and the locations of the magnetic poles and equator. The magnetic equator in the Atlantic sector has moved northward as much as 15° in the past century, and the equatorial ionization anomaly moved correspond- ingly. The electrical conductivity of the ionosphere depends on the strength and direction of the geomagnetic field [e.g., Takeda, 1996], so the ionospheric dynamo and consequent geomagnetic perturbations are affected by field changes. The strength and orientation of the geomagnetic dipole determine how the solar wind and interplanetary magnetic field interact with the magnetosphere, and therefore influence the auroral particle precipitation and high-latitude Joule heating. All of these changes affect the ionosphere [e.g., Cnossen and Richmond, 2008], its interaction with the magnetosphere, and its influence on the thermosphere. In order to understand how observations of long-term trends in the ionosphere relate to thermospheric changes, we need to be able to identify and quantify the effects associated with changes in the geomagnetic field. An additional complexity now appears to have been added to the problem of quantifying anthropogenic change cast against a background of considerable natural variation. During 2008 and 2009, solar activity became extremely low, and the onset of solar cycle 24 was late and weak [e.g., Russell et al., 2010]. A variety of evidence concerning the near-Earth space environment, including ionospheric parameters, radiation belt levels, and thermospheric density, indicated that conditions were significantly quieter than the “usual” solar minimum, and the two-year period between mid-2007 and mid-2009 was one of the longest and quietest solar minima on record. There is evidence from space-based solar measurements that the solar EUV irradiance was also anomalously low during this time, or at least lower than the previous solar minimum [Didkovsky et al., 2010]. And, analyses of satellite orbits show that the upper thermosphere was significantly cooler and less dense than the previous several solar minima, and indeed cooler than any other such period since the beginning of space flight [Emmert et al., 2010]. Figure 2 shows the satellite drag and solar EUV measurements. Model simulations supporting this are shown in Solomon et al. [2010]. Clearly, the 2008-2009 solar minimum was well outside of the expected range due to anthropogenic effects, and solar measurements are commensurate with the atmospheric response. A much smaller change is seen, however, in standard solar proxy indices such as F10.7. Since comparison of successive solar minima was employed by Keating et al. [2000] to make the original detection of thermospheric cooling, and could be a valuable technique for re-

4 moving solar cycle effects, the ramifications of this finding are significant. Quantification of past and future solar EUV levels will be a crucial aspect of understanding anthropogenic global change in the upper atmosphere.

Figure 2. (a) Global mean thermospheric density at 400 km, obtained from satellite orbital parameters over four solar cycles. Blue: 81-day centered running mean. Black: annual average. Green dotted lines: enve-

lope of expected decrease due to increasing CO2 levels, in the range of 2% to 5% per decade, starting with the 1976 annual average. (b) Global mean thermospheric density annual average plotted as a function of the 26-34 nm solar EUV irradiance annual average measured by the SOHO SEM instrument for the ascend- ing (red) and descending (blue) phases of solar cycle 23 [Solomon et al., 2010].

Necessary Measurements In addition to ongoing monitoring of satellite drag, ionospheric parameters, geomagnetic activity indices, and greenhouse gas concentrations, future measurements of the major heating and cooling terms in the upper atmosphere will be required. Solar Inputs The Solar EUV Experiment (SEE) [Woods et al., 1998; 2005] on the TIMED satellite pro- vides measurements in the spectral range 0.1–195 nm, and the SORCE X-ray Photometer System (XPS) augments this data set. Another broad-band data set is provided by the Space Environ- ment Monitor (SEM) on the SOHO spacecraft [Didkovsky et al., 2010] (see above). Because the soft X-ray measurements on TIMED and SORCE are broad-band photometric measurements, similar to the earlier SNOE measurements [Bailey et al., 2000], considerable uncertainty as to their spectral interpretation remains. However, SDO/EVE now measures the solar spectrum from 6 to 106 nm at 0.1 nm resolution with a 10 s cadence [Woods, et al., 2010]. These measurements will be the key to resolving issues in the soft X-ray range, understanding flare irradiance variability, and validating solar spectral irradiance models. In the FUV, measurements by instruments on the UARS satellite [Woods and Rottman, 2002] and the SOLSTICE instrument on board the SORCE satellite [Rottman, 2005, 2006] measure the solar spectrum from H Lyman-alpha (121.6

5 nm) through 300 nm. In addition, the SIM instrument on SORCE covers the range 200 nm through the infrared. Low-resolution instruments on the NOAA GOES satellites now augment this measurement set. It will be extremely important to obtain a comprehensive inter-calibration between TIMED SEE, SOHO/SEM, and the new SDO/EVE data, in order to quantify and understand solar irradiance during the recent solar minimum, and its deviation from past solar minima. For the future, maintaining a continuous program of both high-resolution solar irradiance measurements throughout the spectrum, and low-resolution monitoring (by NOAA GOES) as well, will be crucial to understanding upper atmosphere climate, and also the small but non-negligible solar perturbations of stratosphere and troposphere climate. Infrared Cooling Rates Considerable progress has been made understanding the role of radiatively active minor species in the upper atmosphere, due to satellite observations such as from UARS, CRISTA, SNOE,

TIMED, and others. Global integrated thermospheric cooling rates for CO2 and nitric oxide (NO) obtained from measurements by the TIMED/SABER instrument (updated from the data of Mlynczak et al., 2007) are shown in figure 3. The decline in nitric oxide (NO) cooling during the course of the solar cycle has been successfully simulated, but there are discrepancies between models and measurements of the CO2 cooling rates, as discussed below. These rates are derived from zonal-average rates as a function of latitude and altitude. Infrared measurements will also be valuable for verification of the model altitude profiles of the CO2 15-µm emission and NO 5.3-µm emission, which are critical for climate change studies. Comprehensive future measurements of atmospheric infrared emissions will be necessary to understand the evolution of the terrestrial climate system at all altitudes.

Figure 3. Global integrated thermospheric cooling rates by carbon dioxide and nitric oxide derived from SABER data, over the course of the TIMED mission. Model Development Aeronomical updates to the original Roble and Dickinson [1989] estimates of thermospheric change now produce significantly more NO, in better agreement with observations. In the earlier studies, NO did not enter significantly into the energy budget of the thermosphere and nearly all of the cooling was due to CO2. Current work [Qian et al., 2006; 2008; 2009] indicates significant changes to the predicted global change response, primarily based on increased cooling from the NO 5.3 µm band. Since NO radiation is much larger during solar maximum, NO cooling acts as a regulator, causing the predicted global change to become smaller at solar maximum than at solar minimum due to the decreased importance of CO2 cooling at solar maximum. These estimates agree with the overall rate of change found by Marcos et al. [2005], and exhibit a similar solar cycle dependence found by Emmert et al. [2004], but recent results by Emmert et

6 al. [2008] show even more rapid change, especially at solar minimum. Therefore, having estab- lished the importance of NO cooling, it is necessary to re-evaluate CO2 cooling. The laboratory values for the O-CO2 excitation rate used in the NCAR TIME-GCM and global mean models have historically been lower than those derived from atmospheric measurements or employed in planetary atmosphere models. Global cooling rate data from the TIMED/SABER (c.f., figure 3) agree with model cooling rates for NO, but are considerably higher, about a factor of two, for

CO2. This offers support for the higher collisional excitation rates, and could explain why the model apparently underestimates observed changes. However, changing the model CO2 cooling rate will force a revision of solar heating, conductance, and other parameters controlling the energy budget, in order to retain agreement with observed temperatures and their solar cycle dependence. In addition to a strong upper atmosphere and geospace modeling program, it will be necessary to continue the development of theoretical and empirical models of solar irradiance variability, especially throughout the ultraviolet. Data Analysis Long-term analysis of climate trends will require diverse programs to analyze existing data sources and exploit proposed new measurements. The following overview is not an exhaustive list but describes the scope of the effort. Mesosphere Polar mesospheric cloud observational analysis from the AIM mission, space-based temperature profiles from limb radiometers, and ground-based airglow and lidar observations will be important components of trend analysis and lower-atmosphere interactions. Changes to the hydrogen budget of the mesosphere as a consequence of temperature change and CH4 increases should be monitored. Thermosphere Global-average neutral density derived from satellite drag will continue to be a primary tool for measuring the integrated climate of the thermosphere. New empirical descriptions of the spa- tial distribution of thermospheric density and temperature, based on data assimilation approaches to satellite drag analysis, can augment this. Ultraviolet remote sensing of thermosphere/ionosphere properties is extremely valuable as a global observational tool, and strongly complement the infrared and solar measurements discussed above. Ionosphere Because the ionosphere is much more sensitive to highly variable electromagnetic forces than the thermosphere, resulting in a considerably more complex climatology, the primary focus for ionospheric studies will be to estimate the relative importance of different influences and to understand the underlying physical processes. The predicted responses of the ionospheric NmF2, hmF2, and total electron content (TEC) at different geographic locations to changes in greenhouse gases, solar EUV, geomagnetic activity, and the evolving terrestrial magnetic field need to be modeled and compared with long-term data bases of ground-based ionosonde and radar observations, and future measurements such as from the COSMIC-2 constellation. Exosphere

Calculations of the atomic hydrogen budget, which is a product of CH4 dissociation, predict a large increase in exospheric H, which could ultimately be detectable by ground-based Hα observations. Understanding processes governing the rate of hydrogen escape is important for studies of anthropogenic climate change, but also has implications for climate evolution on geologic time scales. Additionally, changes to H density in the exosphere would effect plasmaspheric H+, potentially ion outflow, and the magnetosphere.

7 References Akmaev, R. A., and V. I. Fomichev, Cooling of the mesosphere and lower thermosphere due to

doubling of CO2. Ann. Geophys.. 16, 1501–1512, 1998. Akmaev, R. A., and V. I. Fomichev, A model estimate of cooling in the mesosphere and lower

thermosphere due to the CO2 increase over the last 3–4 decades. Geophys. Res. Lett.. 27, 2113–2116, 2000. Akmaev, R. A., V. I. Fomichev, and X. Zhu, Impact of middle-atmospheric composition changes on greenhouse cooling in the upper atmosphere. J. Atmos. Terr. Phys.. 68, 1879–1889, 2006. Bailey, S. M., T. N. Woods, C. A. Barth, S. C. Solomon, L. R. Canfield, and R. Korde, Meas- urements of the solar soft X-ray irradiance by the Student Nitric Oxide Explorer: First analysis and underflight calibrations. J. Geophys. Res., 105, 27179, 2000. Beig, G., P. Keckhut, R.P. Lowe, et al., Review of mesospheric temperature trends, Reviews of Geophysics, 41(4), 1015, 2003. Cnossen, I., and A. D. Richmond, Modelling the effects of changes in the Earth’s magnetic field from 1957 to 1997 on the ionospheric hmF2 and foF2 parameters. J. Atmos. Solar-Terr. Phys., 70, 1512-1524, 2008. Dickinson, R. E., E. C. Ridley, and R. G. Roble, A three-dimensional general circulation model of the thermosphere, J. Geophys. Res., 86, 1499-1512, 1981. Didkovsky, L. V., D. L. Judge, and S. R. Wieman, Minima of solar cycles 22/23 and 23/24 as seen in SOHO/CELIAS/SEM absolute solar EUV flux, in ASP Conference Series, SOHO-23: Understanding a Peculiar Solar Minimum, S. R. Cranmer, J. T. Hoeksema, and J. Kohl, eds., Vol. 428, p. 73, 2010. http://arxiv.org/PS_cache/arxiv/pdf/0911/0911.0870v1.pdf. Emmert, J. T., J. M. Picone, J. L. Lean and S. H. Knowles, Compelling evidence of a secular decrease in density, J. Geophys. Res., 109, A02301, 2004. Emmert J. T., J. M. Picone, R. R. Meier, Thermospheric global average density trends, 1967– 2007, derived from orbits of 5000 near-Earth objects, Geophys. Res. Lett., 35, L05101, doi:10.1029/2007GL032809, 2008. Emmert, J. T., J. L. Lean, and J. M. Picone, Observations of record-low thermospheric densities during the 2008 solar minimum, Geophys. Res. Lett., 37, L12102, doi:10.1029/2010GL043671, 2010. Fuller-Rowell, T. J., D. Rees, S. Quegan, R. J. Moffett, and G. J. Bailey, Interactions between neutral thermospheric composition and the polar thermosphere using a coupled global model, J. Geophys. Res., 92, 7744, 1987. Gruzdev, A. N., and G. P. Brasseur, Long-term changes in the mesosphere calculated by a two- dimensional model, J. Geophys. Res., 110, D03304, 2005. Holt, J. M., and S. R. Zhang, Long-term temperature trends in the ionosphere above Millstone Hill, Geophys. Res. Lett., 35, L05813, doi:10.1029/2007GL031148, 2008. Keating, G. M., R. H. Tolson, and M. S. Bradford, Evidence of long-term global decline in the Earth’s thermospheric densities apparently related to anthropogenic effects, Geophys. Res. Lett., 27, 1523-1526, May 15, 2000. Lean, J. L., J. M. Picone, J. T. Emmert, and G. Moore, Thermospheric densities derived from spacecraft orbits: Application to the Starshine satellites, J. Geophys. Res., 111, A04301, doi:10.1029/2005JA011399, 2006. Laštovička, J., R. A. Akmaev, G. Beig, J. Bremer, and J. T. Emmert, Global change in the upper atmosphere, Science, 314, 1253, 2006. Laštovička, J., R. A. Akmaev, G, Beig, J. Bremer, J. T. Emmert, C. Jacobi, M. J. Jarvis, G. Ne- doluha, Y. I. Portnyagin, and T. Ulich, Emerging pattern of global change in the upper atmosphere and ionosphere, Ann. Geophys., 26, 1255, 2008.

8 Marcos, F. A., J. O. Wise, M. J. Kendra, N. J. Grossbard and B. R. Bowman, Detection of a long-term decrease in thermospheric neutral density, Geophys. Res. Lett., 32, L04103, doi:10.1029/2004GL021269, 2005. Mlynczak M. G., F. J. Martin-Torres, B. T. Marshall, R. E. Thompson, J. Williams, T. Turpin, D. P. Kratz, J. M. Russell III, T. N. Woods, and L. L. Gordley, Evidence for a solar cycle influence on the infrared energy budget and radiative cooling of the thermosphere, J. Geophys. Res., 112, A12302, doi:10.1029/2006JA012194, 2007. Qian, L., R. G. Roble, S. C. Solomon, and T. J. Kane, Calculated and observed climate change in the thermosphere, and a prediction for solar cycle 24, Geophys. Res. Lett., 33, L23705, doi:10.1029/2006GL027185, 2006. Qian, L., R. G. Roble, S. C. Solomon, and T. J. Kane, Model simulations of global change in the ionosphere, Geophys. Res. Lett., 35, L07811, doi:10.1029/2007GL033156, 2008. Qian, L., A. G. Burns, S. C. Solomon, and R. G. Roble, The effect of carbon dioxide cooling on trends in the F2-layer ionosphere, J. Atmos. Sol.-Terr. Phys., 71, 1592, doi:10.1016/j.jastp.2009.03.006, 2009. Richmond, A. D., E. C. Ridley, and R. G. Roble, A thermosphere/ionosphere general circulation model with coupled electrodynamics, Geophys. Res. Lett., 19, 601, 1992. Rishbeth, H., and R. G. Roble, Cooling of the upper atmosphere by enhanced greenhouse gases- Modeling of thermospheric and ionospheric effects, Planet. Space Sci., 40, 1011-1026, 1992. Roble, R. G., E. C. Ridley, and R. E. Dickinson, On the global mean structure of the thermosphere, J. Geophys. Res., 92, 8745-8758, 1987. Roble, R. G., E. C. Ridley, A. D. Richmond, and R. E. Dickinson, A coupled thermosphere/ionosphere general circulation model, Geophys. Res. Lett., 15, 1525, 1988. Roble, R. G. and R. E. Dickinson, How will changes in carbon dioxide and methane modify the mean structure of the mesosphere and thermosphere? Geophys. Res. Lett., 16, 1441-1144, 1989. Roble, R. G. and E. C. Ridley, Thermosphere-ionosphere-mesosphere-electrodynamics general circulation model (TIME-GCM): Equinox solar min simulations, 30-500 km, Geophys. Res. Lett., 21, 417, 1994. Rottman, G., The SORCE mission, Solar Physics, 230 (1-2), 7, 2005. Rottman, G., Measurement of total and spectral solar irradiance, Space Sci. Rev., 125, 39, 2006. Russell, C. T., J. G. Luhmann, and L. K. Jian, How unprecedented a solar minimum?, Rev. Geo- phys., 48, RG2004, doi:10.1029/2009RG000316, 2010. Solomon, S. C., T. N. Woods, L. V. Didkovsky, J. T. Emmert, and L. Qian, Anomalously low solar extreme-ultraviolet irradiance and thermospheric density during solar minimum, Geo- phys. Res. Lett., 37, L16103, doi:10.1029/2010GL044468, 2010. Takeda, M., Effects of the strength of the geomagnetic main field strength on the dynamo action in the ionosphere, J. Geophys. Res., 101, 7875, 1996. Woods, T. N., S. M. Bailey, F. Eparvier, G. Lawrence, J. Lean, W. McClintock, R. Roble, G. Rottman, S. C. Solomon, W. K. Tobiska, G. Ucker, and O. R. White, TIMED Solar EUV Experiment, Proc. SPIE, 3442, 180, 1998. Woods, T. N., and G. J. Rottman, Solar ultraviolet variability over time periods of aeronomic interest, in Comparative Aeronomy in the Solar System, AGU Monograph, 2002. Woods, T. N., F. G. Eparvier, S. M. Bailey, P. C. Chamberlain, J. Lean, G. J. Rottman, S. C. Solomon, W. K. Tobiska, and D. Woodraska, The Solar EUV Experiment (SEE): Mission overview and first results, J. Geophys. Res., 110, A01312, doi:10.1029/2004JA010765, 2005. Woods, T. N., et al., Extreme Ultraviolet Variability Experiment (EVE) on the Solar Dynamics Observatory (SDO): Overview of Science Objectives, Instrument Design, Data Products, and Model Developments", Solar Physics, doi: 10.1007/s11207-009-9487-6, 2010.