A Matter of Humidity

Document Sample
A Matter of Humidity Powered By Docstoc


                                                                                                                      How strong a part does water vapor play
       A Matter of Humidity                                                                                           in global warming?

       Andrew E. Dessler1 and Steven C. Sherwood2

               he water vapor feedback is the process              which air leaves stormy regions in a saturated                                  Initial warming influence
               whereby an initial warming of the                   condition, but with negligible ice or liquid
               planet, caused, for example, by an                  content. Water vapor is thereafter transported
                                                                                                                                                                 Thermal radiation
       increase in atmospheric carbon dioxide, leads               by the large-scale circulation, which con-                 12                                  by water vapor
       to an increase in the humidity of the atmo-                 serves the specific humidity (the ratio of the
       sphere. Because water vapor is itself a green-              mass of water vapor to the total mass in a unit

                                                                                                                       Height (km)
       house gas, this increase in humidity causes                 volume of air), except during subsequent sat-
                                                                                                                                                           TEMPERATURE PROFILE
       additional warming. The water vapor feed-                   uration events, when loss of water occurs                                                  AFTER WARMING
       back has long been expected to strongly                     instantaneously to prevent supersaturation.

                                                                                                                                                                                      Downloaded from www.sciencemag.org on February 21, 2009
       amplify climate changes because of the                      Despite the simplicity of this idea, which                                      TEMPERATURE
       expectation that the atmosphere’s relative                  entirely neglects detailed microphysics and                                                             thermal
       humidity would remain roughly constant—                     other small-scale processes, such models                                                               radiation
       meaning that the specific humidity would                    accurately reproduce the observed water                           0
       increase at the rate of the equilibrium vapor               vapor distribution for the mid and upper tro-                         200                            300
       pressure, which rises rapidly with tempera-                 posphere (3, 4). One recent study (5) esti-                                 Temperature (kelvin)
       ture. However, observational evidence has                   mated the uncertainty in the water vapor feed-     Schematic of the water vapor feedback. Because
       been harder to come by, and the effect has                  back associated with microscale process            thermal emission to space by water vapor does not
       been controversial. Much of that controversy                behavior at less than 5%, as a result of the       increase, more warming is needed to balance a
       can now be laid to rest, thanks to new observa-             overwhelming control of humidity by the            given energy input.
       tions and better theoretical understanding.                 large-scale wind field.
           In the 1990s, there was little observational                Thus, the water vapor feedback is essen-       feedbacks is also smaller than the water vapor
       or theoretical understanding of atmospheric                 tially controlled by the large-scale dynamics      feedback—about half the magnitude—and
       humidity and how it varied with global cli-                 and the saturation specific humidity in the        more consistent among climate models (8),
       mate. As a result, debate raged over whether                outflow of the tropical deep convective sys-       because no model predicts substantial and
       the water vapor feedback would really occur,                tems. Convective outflow temperature should,       systematic changes in relative humidity.
       with some very influential proposals that it                on average, warm along with the mean atmo-             Despite these advances, observational evi-
       would not (1). In particular, many believed                 sphere, thus producing the feedback (6, 7).        dence is crucial to determine whether models
       that atmospheric humidity and the water                         Given these considerations, there are good     really capture the important aspects of the
       vapor feedback were controlled by pro-                      reasons to expect global climate models to         water vapor feedback. Such evidence is now
       cesses—such as the details of cloud dynamics                accurately simulate the water vapor feedback:      available from satellite observations of the
       and microphysical processes—that are not                    The large-scale wind and temperature fields        response of atmospheric humidity (and its
       sufficiently well understood and inadequately               that mainly control the humidity are explicitly    impacts on planetary radiation) to a number of
       represented in climate models.                              calculated from the basic fluid equations,         climate variations. Observations during the
           Successive reports from the Intergovern-                unlike small-scale processes that must be rep-     seasonal cycle, the El Niño cycle, the sudden
       mental Panel on Climate Change (IPCC)                       resented by crude parameterizations.               cooling after the 1991 eruption of Mount Pina-
       have suggested increasing confidence in our                     Although the water vapor feedback is           tubo, and the gradual warming over recent
       understanding of the water vapor feedback,                  strong in all global climate models, its magni-    decades all show atmospheric humidity chang-
       but they have remained cautious in defending                tude varies somewhat due to differences            ing in ways consistent with those predicted by
       its magnitude. However, recent advances                     among the models in the amount of upper tro-       global climate models, implying a strong and
       have placed the traditional view of the water               pospheric warming (and hence the increase in       positive water vapor feedback (9–13). A strong
       vapor feedback on a stronger footing than is                specific humidity) per unit of surface warm-       and positive water vapor feedback is also nec-
       widely appreciated.                                         ing. The spread among models in the water          essary for models to explain the magnitude of
           The water vapor feedback mainly results                 vapor feedback is, however, largely compen-        past natural climate variations (14).
       from changes in humidity in the tropical upper              sated by an opposite spread in the “lapse-rate         Both observations and models suggest that
       troposphere (2), where temperatures are far                 feedback,” a negative feedback that occurs         the magnitude of the water vapor feedback is
       below that of the surface and the vapor is                  because a warmer atmosphere radiates more          similar to that obtained if the atmosphere held
       above most of the cloud cover. The distribu-                power to space, thereby reducing net surface       relative humidity constant everywhere. This
       tion of humidity in this region is well repro-              warming. As a result, the sum of the two feed-     should not be taken to mean that relative
       duced by “large-scale control” models, in                   backs is insensitive to errors in predicted        humidity will remain exactly the same every-
                                                                   warming of the upper troposphere, and to           where. Regional variations of relative humid-
                                                                   quantify the sum accurately, one only needs to     ity are seen in all observed climate variations
       1Texas A&M University, College Station, TX 77843, USA.      know how relative humidity (the ratio of spe-      and in model simulations of future climate,
       E-mail: adessler@tamu.edu. Climate Change Research
       Centre, University of New South Wales, Sydney, Australia.   cific humidity to that in a saturated condition)   but have a negligible net impact on the global
       E-mail: s.sherwood@unsw.edu.au                              changes as the climate warms. The sum of the       feedback (12).

1020                                               20 FEBRUARY 2009           VOL 323      SCIENCE      www.sciencemag.org
                                                                                   Published by AAAS

                                                                                                                                                                7. K. Minschwaner, A. E. Dessler, J. Climate 17, 1272
                                      Thus, although there continues to be some              uncertainties in our simulations of the climate,                      (2004).
                                  uncertainty about its exact magnitude, the                 but evidence for the water vapor feedback—                         8. B. J. Soden, I. M. Held, J. Climate 19, 3354 (2006).
                                  water vapor feedback is virtually certain to be            and the large future climate warming it                            9. B. J. Soden, R. T. Wetherald, G. L. Stenchikov, A. Robok,
                                                                                                                                                                   Science 296, 727 (2002).
                                  strongly positive, with most evidence support-             implies—is now strong.                                            10. P. M. D. Forster, M. Collins, Climate Dyn. 23, 207 (2004).
                                  ing a magnitude of 1.5 to 2.0 W/m2/K, suffi-                                                                                 11. B. J. Soden, D. L. Jackson, V. Ramaswamy, M. D.
                                  cient to roughly double the warming that                        References                                                       Schwarzkopf, X. Huang, Science 310, 841 (2005).
                                                                                               1. R. S. Lindzen, Bull. Am. Meteor. Soc. 71, 288 (1990).        12. A. E. Dessler, P. Yang, Z. Zhang, Geophys. Res. Lett. 35,
                                  would otherwise occur. To date, observational                                                                                    L20704 (2008).
                                                                                               2. I. M. Held, B. J. Soden, Ann. Rev. Energy Environ. 25,
                                  records are too short to pin down the exact size                441 (2000).                                                  13. A. K. Inamdar, V. Ramanathan, J. Geophys. Res. 103,
                                  of the water vapor feedback in response to                   3. R. T. Pierrehumbert, H. Brogniez, R. Roca, in The Global         32177 (1998).
                                                                                                  Circulation of the Atmosphere, T. Schneider, A. H. Sobel,    14. A. Hall, S. Manabe, J. Climate 12, 2327 (1999).
                                  long-term warming from anthropogenic                                                                                         15. We thank T. Schneider, P. O’Gorman, and P. Forster for
                                                                                                  Eds. (Princeton Univ. Press, Princeton, 2007),
                                  greenhouse gases. However, it seems unlikely                    pp. 143–218.                                                     their comments on this perspective. This work was sup-
                                  that the water vapor feedback in response to                 4. I. Folkins, K. K. Kelly, E. M. Weinstock, J. Geophys. Res.       ported by NASA grant NNX08AR27G to Texas A&M
                                  long-term warming would behave differently                      107, 4736 (2002).                                                University.
                                                                                               5. S. C. Sherwood, C. L. Meyer, J. Climate 19, 6278 (2006).
                                  from that observed in response to shorter-time               6. D. L. Hartmann, K. Larson, Geophys. Res. Lett. 29, 1951
                                  scale climate variations. There remain many                     (2002).                                                                                         10.1126/science.1171264

                                                                                                                                                                                                                                   Downloaded from www.sciencemag.org on February 21, 2009
                                  CELL BIOLOGY

                                                                                                                                                               Transcriptional regulators that respond
                                  Stress Response and Aging                                                                                                    to stress also influence life span.

                                  Laura R. Saunders and Eric Verdin

                                            xposure to a variety of mild                                                                                                part by increasing SIRT1 expression,
                                                                                                    AC            AC          AC              AC     AC AC AC
                                            stressors, including calorie                                                                                                and in yeast, worms, and fruit flies, the
                                            restriction, thermal stress, or                          p53         NF-κB          HSF1          FOXO1 3 4      PGC-1α     lack of Sir2 abrogates the effects of
                                  hyperbaric oxygen, induces an ad-                                                                                                     calorie restriction on life span (6).
                                  aptive biological response that in-                    SIRT1          Calorie restriction; oxidative stress; resveratrol              Similarly, mice lacking SIRT1 do not
                                  creases eukaryotic life span (1).                                                                                                     show some of the beneficial effects of
                                  There are also a variety of mutations                              p53         NF-κB           HSF1         FOXO1 3 4      PGC-1α     calorie restriction related to longevity
                                  associated with both increased resist-               AC                                                                               (7, 8). The enzymatic activity of
                                  ance to stress and increased longe-                             Replicative                   Protein                                 SIRT1 is activated by resveratrol, a
                                                                                                  senescence Inflammation       homeostasis
                                  vity, such as those associated with                                                                                                   polyphenol produced by plants under
                                                                                                         Apoptosis                    Stress resistance
                                  altered insulin/IGF1 (insulin-like                                                                                                    stress. Resveratrol extends the life-
                                  growth factor 1) signaling in the nem-                                                                                                span of yeast, worms, and flies only
                                  atode Caenorhabditis elegans (2).                                                 Life-span regulation                                when Sir2 is present (9).
                                  Adaptive responses to stressors are Handling stress. SIRT1 is a deacetylase that is activated by a variety of                            Westerheide et al. show that in
                                  mediated by transcription factors that stressors and targets transcriptional regulators including p53, NF-κB, mammalian cells, SIRT1 directly de-
                                  regulate both stress response and life HSF1, FOXO1, 3, and 4, and PGC-1α. These factors then control adaptive acetylates HSF1 and thereby regulates
                                  span. On page 1063 of this issue, responses that modulate life span. AC, acetyl group.                                                the heat shock response. The effect of
                                  Westerheide et al. (3) connect two                                                                                                    SIRT1 on HSF1 appears to be dynam-
                                  additional transcriptional regulators to stress tion—HSF1 trimerizes, translocates to the ically regulated. In response to stresses,
                                  responses and longevity. The results support nucleus, becomes phosphorylated, and binds including heat shock, HSF1 is acetylated by
                                  the idea that low levels of stressors influence to regulatory elements (promoters) of genes the histone acetyltransferase p300, a modifi-
                                  life span and provide additional potential that encode heat shock proteins (4). Heat cation that is thought to function as an “off ”
                                  molecular targets that can be further manipu- shock proteins such as Hsp70 serve as chaper- signal by triggering the dissociation of HSF1
                                  lated experimentally or therapeutically.                     ones and proteases that resolve damaged, mis- from its target gene. Inhibiting SIRT1 expres-
                                      Westerheide et al. demonstrate that the folded, and aggregated proteins.                                                  sion via small interfering RNA prevents HSF1
                                  activity of a transcription factor called heat                   SIRT1, a mammalian ortholog of the yeast from binding to the hsp70 promoter and
                                  shock factor 1 (HSF1) is regulated by the transcriptional regulator Sir2, is a stress- suppresses transcription of the gene when
                                  enzyme sirtuin 1 (SIRT1). HSF1 exists as a activated nicotinamide adenine dinucleotide cells are exposed to heat shock. Conversely,
                                  monomer in unstressed mammalian cells. In (NAD+)–dependent protein deacetylase that Westerheide et al. observed that SIRT1 activa-

                                  response to a variety of stresses—including regulates cell survival, replicative senescence, tion by resveratrol or SIRT1 overexpression
                                  heat shock, hypoxia, misfolded proteins, free inflammation, and metabolism through the in cells decreases HSF1 acetylation, pro-
                                  radicals, and adenosine triphosphate deple- deacetylation of histones (the major protein longs HSF1 binding to target promoters, and
                                                                                               components of chromatin) and other cellular enhances the heat shock response.
                                                                                               factors including the transcription factors p53,                     HSF1 is acetylated on at least nine lysine
                                   Gladstone Institute of Virology and Immunology, 1650        NF-κB, and FOXO1, 3, and 4, and the tran- residues. One acetylated residue, Lys80, con-
                                   Owens Street, San Francisco, CA 94158, USA, and
                                   University of California, San Francisco, CA 94158, USA.     scriptional regulator PGC-1α (5) (see the fig- trols HSF binding to DNA. Thus, acetylation
                                   E-mail: everdin@gladstone.ucsf.edu                          ure). Calorie restriction extends life span in of HSF1 at Lys80 may cause the release of

                                                                             www.sciencemag.org               SCIENCE           VOL 323          20 FEBRUARY 2009                                                               1021
                                                                                                                  Published by AAAS

Shared By: