• Anderson, P. S., 1995: Mechanism for the behavior of hydroactive materials used in humidity sensors. J. Atmos. Oceanic Technol., 12, 662667, https://doi.org/10.1175/1520-0426(1995)012<0662:MFTBOH>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Andreas, E. L., P. S. Guest, P. O. G. Persson, C. W. Fairall, T. W. Horst, R. E. Moritz, and S. R. Semmer, 2002: Near-surface water vapor over polar sea ice is always near ice saturation. J. Geophys. Res., 107, 8033, https://doi.org/10.1029/2000JC000411.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Boisvert, L. N., and J. C. Stroeve, 2015: The Arctic is becoming warmer and wetter as revealed by the Atmospheric Infrared Sounder. Geophys. Res. Lett., 42, 44394446, https://doi.org/10.1002/2015GL063775.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bring, A., and Coauthors, 2016: Arctic terrestrial hydrology: A synthesis of processes, regional effects, and research challenges. J. Geophys. Res. Biogeosci., 121, 621649, https://doi.org/10.1002/2015JG003131.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Brunke, M. A., S. T. Stegall, and X. Zeng, 2015: A climatology of tropospheric humidity inversions in five reanalyses. Atmos. Res., 153, 165187, https://doi.org/10.1016/j.atmosres.2014.08.005.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Curry, J., 1983: On the formation of continental polar air. J. Atmos. Sci., 40, 22782292, https://doi.org/10.1175/1520-0469(1983)040<2278:OTFOCP>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dee, D. P., and Coauthors, 2011: The ERA‐Interim reanalysis: Configuration and performance of the data assimilation system. Quart. J. Roy. Meteor. Soc., 137, 553597, https://doi.org/10.1002/qj.828.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Devasthale, A., J. Sedlar, and M. Tjernström, 2011: Characteristics of water-vapour inversions observed over the Arctic by Atmospheric Infrared Sounder (AIRS) and radiosondes. Atmos. Chem. Phys., 11, 98139823, https://doi.org/10.5194/acp-11-9813-2011.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dufour, A., O. Zolina, and S. K. Gulev, 2016: Atmospheric moisture transport to the Arctic: Assessment of reanalyses and analysis of transport components. J. Climate, 29, 50615081, https://doi.org/10.1175/JCLI-D-15-0559.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Durre, I., and X. Yin, 2008: Enhanced radiosonde data for studies of vertical structure. Bull. Amer. Meteor. Soc., 89, 12571262.

  • Durre, I., R. S. Vose, and D. B. Wuertz, 2006: Overview of the integrated global radiosonde archive. J. Climate, 19, 5368, https://doi.org/10.1175/JCLI3594.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Francis, J. A., D. M. White, J. J. Cassano, W. J. Gutowski Jr., L. D. Hinzman, M. M. Holland, M. A. Steele, and C. J. Vörösmarty, 2009: An Arctic hydrologic system in transition: Feedbacks and impacts on terrestrial, marine, and human life. J. Geophys. Res., 114, G04019, https://doi.org/10.1029/2008JG000902.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gettelman, A., V. P. Walden, L. M. Miloshevich, W. L. Roth, and B. Halter, 2006: Relative humidity over Antarctica from radiosondes, satellites, and a general circulation model. J. Geophys. Res., 111, D09S13, https://doi.org/10.1029/2005JD006636.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ingleby, B., 2017: An assessment of different radiosonde types 2015/2016. ECMWF Tech. Memo. 807, 69 pp., https://www.ecmwf.int/sites/default/files/elibrary/2017/17551-assessment-different-radiosonde-types-20152016.pdf.

  • Intrieri, J. M., C. W. Fairall, M. D. Shupe, P. O. G. Persson, E. L. Andreas, P. S. Guest, and R. E. Moritz, 2002a: An annual cycle of Arctic surface cloud forcing at SHEBA. J. Geophys. Res., 107, 8039, https://doi.org/10.1029/2000JC000439.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Intrieri, J. M., M. D. Shupe, T. Uttal, and B. J. McCarty, 2002b: An annual cycle of Arctic cloud characteristics observed by radar and lidar at SHEBA. J. Geophys. Res., 107, 8030, https://doi.org/10.1029/2000JC000423.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Jakobson, E., and T. Vihma, 2010: Atmospheric moisture budget in the Arctic based on the ERA-40 reanalysis. Int. J. Climatol., 30, 21752194, https://doi.org/10.1002/joc.2039.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Jakobson, E., T. Vihma, T. Palo, L. Jakobson, H. Keernik, and J. Jaagus, 2012: Validation of atmospheric reanalyses over the central Arctic Ocean. Geophys. Res. Lett., 39, L10802, https://doi.org/10.1029/2012GL051591.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kobayashi, S., and Coauthors, 2015: The JRA-55 reanalysis: General specifications and basic characteristics. J. Meteor. Soc. Japan, 93, 548, https://doi.org/10.2151/jmsj.2015-001.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lüpkes, C., T. Vihma, E. Jakobson, G. König‐Langlo, and A. Tetzlaff, 2010: Meteorological observations from ship cruises during summer to the central Arctic: A comparison with reanalysis data. Geophys. Res. Lett., 37, L09810, https://doi.org/10.1029/2010GL042724.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Miloshevich, L. M., H. Vömel, D. N. Whiteman, B. M. Lesht, F. J. Schmidlin, and F. Russo, 2006: Absolute accuracy of water vapor measurements from six operational radiosonde types launched during AWEX‐G and implications for AIRS validation. J. Geophys. Res., 111, D09S10, https://doi.org/10.1029/2005JD006083.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Nygård, T., T. Valkonen, and T. Vihma, 2013: Antarctic low-tropospheric humidity inversions: 10-yr climatology. J. Climate, 26, 52055219, https://doi.org/10.1175/JCLI-D-12-00446.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Nygård, T., T. Valkonen, and T. Vihma, 2014: Characteristics of Arctic low-tropospheric humidity inversions based on radio soundings. Atmos. Chem. Phys., 14, 19591971, https://doi.org/10.5194/acp-14-1959-2014.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Persson, P. O. G., C. W. Fairall, E. L. Andreas, P. S. Guest, and D. K. Perovich, 2002: Measurements near the Atmospheric Surface Flux Group tower at SHEBA: Near‐surface conditions and surface energy budget. J. Geophys. Res., 107, 8045, https://doi.org/10.1029/2000JC000705.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Prata, A. J., 1996: A new long-wave formula for estimating downward clear-sky radiation at the surface. Quart. J. Roy. Meteor. Soc., 122, 11271151, https://doi.org/10.1002/qj.49712253306.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Savré, J., A. M. L. Ekman, G. Svensson, and M. Tjernström, 2015: Large-eddy simulations of an Arctic mixed-phase stratiform cloud observed during ISDAC: Sensitivity to moisture aloft, surface fluxes and large-scale forcing. Quart. J. Roy. Meteor. Soc., 141, 11771190, https://doi.org/10.1002/qj.2425.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Seager, R., and N. Henderson, 2013: Diagnostic computation of moisture budgets in the ERA-Interim reanalysis with reference to analysis of CMIP-archived atmospheric model data. J. Climate, 26, 78767901, https://doi.org/10.1175/JCLI-D-13-00018.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sedlar, J., 2014: Implications of limited liquid water path on static mixing within Arctic low-level clouds. J. Appl. Meteor. Climatol., 53, 27752789, https://doi.org/10.1175/JAMC-D-14-0065.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sedlar, J., M. D. Shupe, and M. Tjernström, 2012: On the relationship between thermodynamic structure and cloud top, and its climate significance in the Arctic. J. Climate, 25, 23742393, https://doi.org/10.1175/JCLI-D-11-00186.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Serreze, M. C., and Coauthors, 2006: The large-scale freshwater cycle of the Arctic. J. Geophys. Res., 111, C11010, https://doi.org/10.1029/2005JC003424.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Shupe, M. D., and J. M. Intrieri, 2004: Cloud radiative forcing of the Arctic surface: The influence of cloud properties, surface albedo, and solar zenith angle. J. Climate, 17, 616628, https://doi.org/10.1175/1520-0442(2004)017<0616:CRFOTA>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Shupe, M. D., P. O. G. Persson, I. M. Brooks, M. Tjernström, J. Sedlar, T. Mauritsen, S. Sjogren, and C. Leck, 2013: Cloud and boundary layer interactions over the Arctic sea ice in late summer. Atmos. Chem. Phys., 13, 93799400, https://doi.org/10.5194/acp-13-9379-2013.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Solomon, A., M. D. Shupe, P. O. G. Persson, and H. Morrison, 2011: Moisture and dynamical interactions maintaining decoupled Arctic mixed-phase stratocumulus in the presence of a humidity inversion. Atmos. Chem. Phys., 11, 10 12710 148, https://doi.org/10.5194/acp-11-10127-2011.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Solomon, A., M. D. Shupe, P. O. G. Persson, H. Morrison, T. Yamaguchi, P. M. Caldwell, and G. de Boer, 2014: The sensitivity of springtime Arctic mixed-phase stratocumulus clouds to surface-layer and cloud-top inversion-layer moisture sources. J. Atmos. Sci., 71, 574595, https://doi.org/10.1175/JAS-D-13-0179.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sotiropoulou, G., and Coauthors, 2016: Atmospheric conditions during the Arctic Clouds in Summer Experiment (ACSE): Contrasting open water and sea ice surfaces during melt and freeze-up seasons. J. Climate, 29, 87218744, https://doi.org/10.1175/JCLI-D-16-0211.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Stramler, K., A. D. Del Genio, and W. B. Rossow, 2011: Synoptically driven Arctic winter states. J. Climate, 24, 17471762, https://doi.org/10.1175/2010JCLI3817.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Susskind, J., J. M. Blaisdell, and L. Iredell, 2014: Improved methodology for surface and atmospheric soundings, error estimates, and quality control procedures: The Atmospheric Infrared Sounder science team version-6 retrieval algorithm. J. Appl. Remote Sens., 8, 084994, https://doi.org/10.1117/1.JRS.8.084994.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Tastula, E.-M., T. Vihma, E. L. Andreas, and B. Galperin, 2013: Validation of the diurnal cycles in atmospheric reanalyses over Antarctic sea ice. J. Geophys. Res. Atmos., 118, 41944204, https://doi.org/10.1002/jgrd.50336.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Tjernström, M., 2005: The summer Arctic boundary layer during the Arctic Ocean Experiment 2001 (AOE-2001). Bound.-Layer Meteor., 117, 536, https://doi.org/10.1007/s10546-004-5641-8.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Tjernström, M., and Coauthors, 2012: Meteorological conditions in the central Arctic summer during the Arctic Summer Cloud Ocean Study (ASCOS). Atmos. Chem. Phys., 12, 68636889, https://doi.org/10.5194/acp-12-6863-2012.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Tjernström, M., and Coauthors, 2015: Warm‐air advection, air mass transformation and fog causes rapid ice melt. Geophys. Res. Lett., 42, 55945602, https://doi.org/10.1002/2015GL064373.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Vihma, T., T. Kilpeläinen, M. Manninen, A. Sjöblom, E. Jakobson, T. Palo, J. Jaagus, and M. Maturilli, 2011: Characteristics of temperature and humidity inversions and low-level jets over Svalbard fjords in spring. Adv. Meteor., 2011, 486807, https://doi.org/10.1155/2011/486807.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Vihma, T., and Coauthors, 2016: The atmospheric role in the Arctic water cycle: A review on processes, past and future changes, and their impacts. J. Geophys. Res. Biogeosci., 121, 586620, https://doi.org/10.1002/2015JG003132.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wesslén, C., M. Tjernström, D. H. Bromwich, G. de Boer, A. M. L. Ekman, L.-S. Bai, and S.-H. Wang, 2014: The Arctic summer atmosphere: An evaluation of reanalyses using ASCOS data. Atmos. Chem. Phys., 14, 26052624, https://doi.org/10.5194/acp-14-2605-2014.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Woods, C., R. Caballero, and G. Svensson, 2013: Large‐scale circulation associated with moisture intrusions into the Arctic during winter. Geophys. Res. Lett., 40, 47174721, https://doi.org/10.1002/grl.50912.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, T., K. Stamnes, and S. A. Bowling, 2001: Impact of the atmospheric thickness on the atmospheric downwelling longwave radiation and snowmelt under clear-sky conditions in the Arctic and Subarctic. J. Climate, 14, 920939, https://doi.org/10.1175/1520-0442(2001)014<0920:IOTATO>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • View in gallery

    The division of the study regions and the sounding stations used. The regions include the following sounding stations: Arctic Ocean: Tiksi Bay (14), Barrow (29), Eureka (34); northern Atlantic: Jan Mayen (1), Ny Ålesund (2), Bear Island (4); northern Europe: Bodø (3), Luleå-Kallax (5), Sodankylä (6), Malye Karmakuly (12), Murmansk (16), Kandalaksa (17), Shoina (18), Narian Mar (19), Pechora (21); Siberia: Ostrov Dikson (11), Khatanga (13), Tiksi Bay (14), Cokurdah (15), Sale-Khard (Obdorsk) (20), Turuhansk (22), Olenek (23), Verkhoyansk (24), Gigansk (25), Cherskiy (26), Zyrianka (27), Omolon (28); Alaska/Canada: Barrow (29), Kotzebue (30), Norman Wells (31), Hall Beach (32), Alert (33), Eureka (34), Resolute Bay (35), Cambridge Bay (36); Greenland: Egedesminde (7), Danmarkshavn (8), Scoresbysund (9), Ammassalik (10).

  • View in gallery

    (a) Example of SHI (below the 950-hPa level) caused by surface cooling and moisture condensation and (b) example of SHIs caused by humidity advection. The red lines are temperature, blue lines are dewpoint temperature, and thin dashed lines are specific humidity isolines. SHIs are the layers where the dewpoint temperature curves are tilted clockwise from the specific humidity isolines (dashed lines).

  • View in gallery

    Schematic illustration of processes leading or contributing to the formation of SHIs.

  • View in gallery

    (left) DJF mean specific humidity from ERA-Interim, JRA-55, and soundings. (center) Bias and (right) root-mean-square difference of ERA-Interim (dashed line) and JRA-55 (solid line) in the comparison with soundings in winter.

  • View in gallery

    As in Fig. 4, but for JJA.

  • View in gallery

    DJF and JJA means of SHI strength for the locations of the sounding stations from ERA-Interim (solid line), JRA-55 (long dashed line), and soundings (short dashed lines).

  • View in gallery

    DJF and JJA means of SHI occurrence in 100-hPa-thick layers for the study regions from ERA-Interim and JRA-55 (solid lines) and for the locations of the sounding stations from ERA-Interim, JRA-55, and soundings (dashed lines).

  • View in gallery

    (a) DJF and (b) JJA means of vertically integrated water vapor from ERA-Interim and soundings (circles). DJF and JJA means of specific humidity cross sections along latitudes (c),(d) 65°N and (e),(f) 75°N from ERA-Interim.

  • View in gallery

    DJF and JJA means of the fraction of SHIs that occurred simultaneously with TIs in 50-hPa-thick layers for the study regions from ERA-Interim and JRA-55 (solid lines) and for the locations of the sounding stations from ERA-Interim, JRA-55, and soundings (dashed lines).

  • View in gallery

    As in Fig. 9, but for SHIs that occurred simultaneously with RHIs.

  • View in gallery

    (a),(b) DJF and (c),(d) JJA means of SHI occurrence between the level of 800 hPa and the surface from ERA-Interim, JRA-55, and soundings (circles).

  • View in gallery

    Monthly means of SHI strength from ERA-Interim.

  • View in gallery

    The contribution of specific humidity advection to the increase of SHI strength on layers between subsequent model levels below the 800-hPa level based on ERA-Interim.

  • View in gallery

    (a),(b) DJF and (c),(d) JJA means of SHI strength from ERA-Interim, JRA-55, and soundings (circles).

  • View in gallery

    (a),(c),(e) SHI strength and moisture flux density at the level of 850 hPa (black arrows) from ERA-Interim. (b),(d),(f) Specific humidity cross section along longitude 150°E [marked with a red line in (a), (c), and (e)] and meridional moisture flux density (black arrows) from ERA-Interim.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 41 41 28
PDF Downloads 35 35 20

Arctic Humidity Inversions: Climatology and Processes

View More View Less
  • 1 Finnish Meteorological Institute, Helsinki, Finland
© Get Permissions
Full access

Abstract

The occurrence and characteristics of Arctic specific humidity inversions (SHIs) were examined on the basis of two reanalyses (ERA-Interim and JRA-55) and radiosonde sounding data from 2003 to 2014. Based on physical properties, the SHIs were divided into two main categories: SHIs below and above the 800-hPa level. Above the 800-hPa level, SHIs occurred simultaneously with relative humidity inversions and without the presence of a temperature inversion; these SHIs were probably formed when a moist air mass was advected over a dry air mass. SHIs below the 800-hPa level occurred simultaneously with temperature inversions in conditions of high relative humidity, which suggests that condensation had an important role in SHI formation. Below the 800-hPa level, SHI occurrence had a large seasonal and spatial variation, which depended on the surface heat budget. In winter, most SHIs were formed because of surface radiative cooling, and the occurrence of SHIs was high (even exceeding 90% of the time) on continents and over the ice-covered Arctic Ocean. In summer, the occurrence of SHIs was highest (70%–90%) over the coastal Arctic Ocean, where SHIs were generated by warm and moist air advection over a cold sea surface. In the reanalyses, the strongest SHIs occurred in summer over the Arctic Ocean. The comparisons between radiosonde soundings and the reanalyses showed that the main features of the seasonal and spatial variation of SHI occurrence and SHI strength were well represented in the reanalyses, but SHI strength was underestimated.

© 2018 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

Corresponding author: Tuomas Naakka, tuomas.naakka@fmi.fi

Abstract

The occurrence and characteristics of Arctic specific humidity inversions (SHIs) were examined on the basis of two reanalyses (ERA-Interim and JRA-55) and radiosonde sounding data from 2003 to 2014. Based on physical properties, the SHIs were divided into two main categories: SHIs below and above the 800-hPa level. Above the 800-hPa level, SHIs occurred simultaneously with relative humidity inversions and without the presence of a temperature inversion; these SHIs were probably formed when a moist air mass was advected over a dry air mass. SHIs below the 800-hPa level occurred simultaneously with temperature inversions in conditions of high relative humidity, which suggests that condensation had an important role in SHI formation. Below the 800-hPa level, SHI occurrence had a large seasonal and spatial variation, which depended on the surface heat budget. In winter, most SHIs were formed because of surface radiative cooling, and the occurrence of SHIs was high (even exceeding 90% of the time) on continents and over the ice-covered Arctic Ocean. In summer, the occurrence of SHIs was highest (70%–90%) over the coastal Arctic Ocean, where SHIs were generated by warm and moist air advection over a cold sea surface. In the reanalyses, the strongest SHIs occurred in summer over the Arctic Ocean. The comparisons between radiosonde soundings and the reanalyses showed that the main features of the seasonal and spatial variation of SHI occurrence and SHI strength were well represented in the reanalyses, but SHI strength was underestimated.

© 2018 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

Corresponding author: Tuomas Naakka, tuomas.naakka@fmi.fi

1. Introduction

Water vapor is an important component of the Arctic climate system (Serreze et al. 2006; Francis et al. 2009; Vihma et al. 2016). Water vapor is present in the Arctic atmosphere because of local evaporation and evapotranspiration (Bring et al. 2016; Boisvert and Stroeve 2015) and transport from lower latitudes, most of which is carried by synoptic-scale cyclones (Jakobson and Vihma 2010; Dufour et al. 2016). Under clear skies, water vapor enhances the atmospheric emissivity for longwave radiation, seen as increased downward longwave radiation at Earth’s surface (Prata 1996; Zhang et al. 2001; Devasthale et al. 2011). The increase in the atmospheric emissivity is even larger when water vapor condenses into cloud or fog droplets (Shupe and Intrieri 2004). Arctic clouds have a warming effect on the surface during most of the year because their effect of increasing the downward longwave radiation dominates their effect of reducing the net solar radiation over high-albedo snow and ice surfaces. In summer, however, clouds typically have a cooling effect on surface types with a lower albedo, such as the open sea, melting sea ice, and ground (Intrieri et al. 2002a; Shupe and Intrieri 2004).

The altitude where water vapor condenses to clouds or fog depends on the vertical profiles of air temperature and specific humidity. On the global scale, a typical situation is that both air temperature and specific humidity decrease with height. In the Arctic, however, the lower troposphere often includes inversion layers where the temperature or specific humidity or both increase with height (Devasthale et al. 2011; Sedlar et al. 2012; Shupe et al. 2013; Nygård et al. 2014; Brunke et al. 2015). A specific humidity inversion (SHI) can contribute to cloud formation and maintenance. An SHI above a cloud layer can provide an adequate moisture source for Arctic stratocumulus clouds, and this feature probably allows for the maintenance of extensive cloudiness in summer (Solomon et al. 2011, 2014; Sedlar et al. 2012; Sedlar 2014). If a cloud is decoupled from Earth’s surface, an SHI may be the only moisture source for the cloud (Sedlar et al. 2012; Savré et al. 2015).

SHIs often occur simultaneously with temperature inversions (TIs) (Andreas et al. 2002; Persson et al. 2002; Vihma et al. 2011; Tjernström et al. 2012; Sotiropoulou et al. 2016). In winter, surface cooling related to the negative net radiation leads to the formation of TIs and increases the relative humidity, and further leads to the formation of SHIs via moisture condensation (Curry 1983). Therefore, the properties of SHIs and TIs are partly connected. Devasthale et al. (2011) and Nygård et al. (2014) found that the strength of SHIs and TIs are often linked. Nygård et al. (2014) also found that the strength, base height, and depth of the strongest SHI and TI in each sounding profile were correlated. Another mechanism affecting the formation of SHIs is humidity advection. Nygård et al. (2014) showed that approximately half of all Arctic SHIs occurred without the presence of TIs, and Nygård et al. (2013) showed that vertical changes in humidity advection, and especially the near-surface advection of dry air by katabatic winds, was an important factor in generating SHIs in the Antarctic. Geographical conditions on the coast of Antarctica are unique and, therefore, it is probable that katabatic processes do not play a major role for Arctic SHIs, except in regions with large slopes, as in Greenland. Brunke et al. (2015) showed that in the Arctic, humidity advection has the largest effect on specific humidity tendencies in SHI layers, whereas moist physics (condensation and evaporation) has a much smaller contribution to the tendencies. However, Brunke et al. (2015) showed that humidity advection has both weakening and strengthening effects on SHI, and they suggested that humidity advection may not be as important a process for the formation of SHIs as the examination of specific humidity tendencies in SHI layers implied.

SHIs typically occur simultaneously on multiple levels (Devasthale et al. 2011; Nygård et al. 2013, 2014). Based on satellite retrievals of the Atmospheric Infrared Sounder (AIRS), Devasthale et al. (2011) showed that in winter under clear skies the occurrence of SHIs exceeds 50% in most of the continental Arctic and the Arctic Ocean, and an even higher SHI occurrence was found in radiosonde sounding data (Devasthale et al. 2011; Nygård et al. 2014). SHI occurrence mostly exceeded 90% at the radiosonde sounding stations north of 65°N (Nygård et al. 2014). Brunke et al. (2015) showed that an SHI is found in wintertime monthly mean specific humidity profiles over the majority of the Arctic based on reanalyses. In summer, the reported occurrence of SHIs has varied considerably between studies. Devasthale et al. (2011) showed a very low, approximately 10%, SHI occurrence in the Arctic in summer based on AIRS satellite retrievals, but notably higher SHI occurrence in the radiosonde sounding observations. Nygård et al. (2014) reported that SHI occurrence in summer was high, on average only 10% lower than in winter. It is noteworthy that the results of AIRS satellite retrievals (Devasthale et al. 2011) were for clear-sky conditions only, and therefore, they were not very representative for the generally cloudy conditions over the Arctic Ocean in summer. In addition, Gettelman et al. (2006) showed that AIRS retrievals are not able to capture very fine details of specific humidity profiles, and the prior information used for AIRS retrievals (Susskind et al. 2014) could affect the low occurrence of SHIs. The strength of SHIs (i.e., the specific humidity difference across the inversion layer) has been found to be larger in summer, even though SHIs are more frequent in winter (Devasthale et al. 2011; Nygård et al. 2014; Brunke et al. 2015).

The regional horizontal distribution of SHI occurrence has been well reported in recent climatological studies (Devasthale et al. 2011; Nygård et al. 2014; Brunke et al. 2015), but the vertical distribution of SHIs and its regional and seasonal variations have remained unstudied. In addition, the strong dependence between the properties of SHIs and TIs and the processes leading to the formation of SHIs have not been well explained so far. In this study, we examine the vertical and regional distributions of SHI occurrence and the regional distributions of SHI strength based on two atmospheric reanalyses and radiosonde soundings. In addition, physical processes behind the formation of SHIs are suggested. The paper is structured as follows: In section 2, the data and calculation methods are presented, and the potential sensitivity of the results to differences between the datasets is discussed. In section 3a, the reanalyses are compared to the radiosonde sounding data to evaluate the accuracy of the reanalyses in representing the atmospheric moisture distribution, and in section 3b, an overview of specific humidity distributions in the Arctic is presented. Then, SHIs above the 800-hPa level (section 3c) and below it (section 3d) are examined separately, because the processes responsible for the formation of SHIs are different between the low troposphere and the atmosphere aloft. In section 4, the key results of this study are compared with previous studies, and the processes behind SHI formation and data accuracy are discussed. The main conclusions of this study are presented in section 5.

2. Study region, material, and methods

We examined the occurrence and strength of SHIs on seasonal time scales and focused on winter (DJF) and summer (JJA), which represent extremes in atmospheric moisture and have clear differences in SHI occurrence and strength, whereas spring and autumn are transition periods between winter- and summer-type SHIs. The study area is the area north of 60°N. The area is divided into six regions based on the spatial patterns of specific humidity conditions in the lower troposphere; the variation of conditions inside each region is much smaller than the variation between regions. These regions are 1) the Arctic Ocean, 2) North Atlantic, 3) northern Europe, 4) Siberia, 5) Alaska and Canada, and 6) Greenland. The regions are outlined in Fig. 1. The study period is 12 years, from January 2003 to December 2014, which is rather short for climatology, but close to the periods addressed by previous studies on Arctic SHIs (Nygård et al. 2014 and Devasthale et al. 2011). The period is short for the evaluation of the trends of SHIs but probably long enough to determine the main features of SHIs and the processes behind their formation.

Fig. 1.
Fig. 1.

The division of the study regions and the sounding stations used. The regions include the following sounding stations: Arctic Ocean: Tiksi Bay (14), Barrow (29), Eureka (34); northern Atlantic: Jan Mayen (1), Ny Ålesund (2), Bear Island (4); northern Europe: Bodø (3), Luleå-Kallax (5), Sodankylä (6), Malye Karmakuly (12), Murmansk (16), Kandalaksa (17), Shoina (18), Narian Mar (19), Pechora (21); Siberia: Ostrov Dikson (11), Khatanga (13), Tiksi Bay (14), Cokurdah (15), Sale-Khard (Obdorsk) (20), Turuhansk (22), Olenek (23), Verkhoyansk (24), Gigansk (25), Cherskiy (26), Zyrianka (27), Omolon (28); Alaska/Canada: Barrow (29), Kotzebue (30), Norman Wells (31), Hall Beach (32), Alert (33), Eureka (34), Resolute Bay (35), Cambridge Bay (36); Greenland: Egedesminde (7), Danmarkshavn (8), Scoresbysund (9), Ammassalik (10).

Citation: Journal of Climate 31, 10; 10.1175/JCLI-D-17-0497.1

a. Datasets

Two modern reanalyses, the European Centre for Medium-Range Weather Forecasts (ECMWF) ERA-Interim (Dee et al. 2011) and the Japan Meteorological Agency (JMA) JRA-55 (Kobayashi et al. 2015), are used for the examination of SHIs in the Arctic. Variables used in this study have been stored in a regular latitude–longitude grid with a horizontal resolution of 0.75° in ERA-Interim and in a Gaussian grid with a horizontal resolution of 0.56° in JRA-55. Both reanalyses have 60 hybrid levels, with very similar vertical resolutions in the troposphere. The model-level fields were used because they have a better vertical resolution in the lower troposphere than the pressure-level fields. Both reanalyses use a four-dimensional variational data assimilation (4D-Var) system, and the analysis fields have been produced with 6-h intervals. As no surface flux data were assimilated into the reanalyses, the products for these variables are only based on short-term forecasts.

The reanalysis products are compared with radiosonde sounding data taken from the Integrated Global Radiosonde Archive (IGRA) (Durre et al. 2006; Durre and Yin 2008). Soundings from 36 stations north of 65°N are used in this study. Quality-assured IGRA sounding data have undergone several quality assurance checks, which have removed outliers from observations, providing a consistent dataset. The quality assurance checks do not remove biases caused, for example, by radiosonde types. IGRA data consist of observed variables and variables derived from the observed variables. Radiosonde soundings are taken mainly twice a day at 0000 and 1200 UTC. Specific humidity has a weak diurnal cycle. Hence, the differences in time resolution have probably only a minimal effect on the differences between the sounding data and the reanalyses. However, in summer over continents, the diurnal cycle of specific humidity may cause a difference between the soundings and reanalyses in SHI occurrence, if no soundings are carried out between 0000 and 0600 local solar time (see the discussion in section 4). The sounding data from the stations located in the six study regions are averaged for each region. As conditions at the sounding stations are possibly not representative for the whole region, the comparisons between the reanalyses and the sounding data are made on the basis of the values at the grid points closest to each sounding station (one grid point per station). The representativeness of the soundings is probably weakest over the oceans because all sounding stations are located on land. Three coastal stations (Eureka, Barrow, and Tiksi Bay), which are located near sea level, were utilized to represent the Arctic Ocean, and stations on small islands, Jan Mayen and Bear Island, and Ny Ålesund were used to represent the Atlantic Ocean.

b. Methods

Our focus was on the occurrence and strength of SHIs, and their spatial and seasonal variation. SHI occurrence is computed by comparing specific humidity values between consecutive model levels up to the 400-hPa level. For analyses, SHI occurrences between individual model levels are combined into thicker layers consisting of several model levels. This method does not yield the probability of SHI occurrence in a single vertical level but the probability of occurrence in the whole layer. SHI occurrence is calculated from the sounding data in the same way as from the reanalyses, except that, instead of model levels, all vertical levels in soundings, that is, mandatory and significant levels are utilized. A vertical resolution of 100 hPa is applied for the analyses of the vertical structure of SHI occurrence. Layers that are 100-hPa thick are used instead of 50-hPa-thick layers because SHI occurrence in the sounding data was systematically higher in the 50-hPa-thick layer above the mandatory pressure levels 850 hPa and 700 hPa than in the 50-hPa-thick layer below these levels. This feature was noticeable at most sounding stations and in addition, at the stations where the upward-decreasing trend of SHI occurrence was strong, the values above a mandatory level were not higher than below the level but deviated similarly from the trend. Those peaks in SHI occurrence are probably caused by the reporting method of sounding variables instead of real physical conditions. The sounding data capture finer details than the reanalyses, especially above the 800-hPa level, which may increase the probability of SHI occurrence in the sounding data. This is because the vertical resolution of reanalyses decreases from nine levels between 900 and 1000 hPa (when the surface pressure is 1000 hPa) to two or three levels per each 100-hPa interval above the 800-hPa level.

SHI strength is computed by subtracting the value of specific humidity at the lowest model level from the vertical maximum value of specific humidity. Therefore, it does not take into account local vertical gradients below the specific humidity maximum, nor the SHIs located above the specific humidity maximum. SHI strength is sensitive to the specific humidity at the lowest model level, which is not at Earth’s surface but at an altitude of approximately 10 m in both reanalyses. Hence, SHI strength does not exactly represent the specific humidity difference between the atmospheric maximum and the surface. When calculating the seasonal-mean SHI strength, zero values from the analysis times when the maximum is located at the lowest model level are included.

The vertical gradient of specific humidity q can be written as follows:
e1
where S is the saturation ratio, ρυ_sat is the saturation density of water vapor, ρ is air density, T is air temperature, and z is the vertical coordinate. The first term on the right-hand side is the effect of the relative humidity vertical gradient on the vertical gradient of specific humidity. The second term on the right-hand side is the effect of saturation vapor density on specific humidity. As the saturation vapor density only depends on temperature, the second term depends on the vertical gradient of temperature. The third term is the effect of the upward decrease in air density on specific humidity. Accordingly, the occurrence of an SHI is related to the occurrence of a relative humidity inversion (RHI), a TI, an upward-decreasing air density, or the occurrence of two or three of these factors simultaneously.

The formation of an SHI can result from humidity advection, condensation, or evaporation. Evaporation from rain or snowfall can generate an SHI only when the dry layer is located below a cloud. Typically the formation of an SHI is connected to moisture condensation or vertically differential humidity advection. In Fig. 2a, the lowest SHI layer (from the surface to 950 hPa), where the dewpoint temperature curve is tilted clockwise from specific humidity isolines, was associated with a cold air mass, and the formation of the SHI was related to moisture condensation due to surface radiative cooling. Surface cooling led to the formation of a TI, and after saturation was reached, further cooling led to moisture condensation and the formation of an SHI (process 3 in Fig. 3). In contrast, the formation of the SHI layer between the 975–950-hPa levels in Fig. 2b was largely contributed by humidity advection, even though the SHI occurred with a TI in saturated conditions (process 2 in Fig. 3). The strong SHI and TI were formed because of temperature and humidity advection due to southwesterly winds. However, the formation of the cold and dry layer near the surface was probably related to radiative surface cooling, and also saturated conditions in the SHI layer suggested the moisture condensation. Accordingly, both processes, humidity advection and condensation, had contributed to the evolution of the SHI. In both Figs. 2a and 2b, SHIs at upper-tropospheric levels occurred with RHIs and unsaturated conditions, hence the formation of these SHIs was caused by humidity advection (process 1 in Fig. 3). In unsaturated conditions, when SHIs occurred simultaneously with RHIs, SHIs were formed because of differential humidity advection, either upward-increasing moist air advection or downward-increasing dry air advection. Moisture condensation or evaporation cannot form or strengthen an SHI when it occurs with an RHI, except in the case of evaporation from rain or condensation in supersaturated conditions when the availability of condensation nuclei determines the occurrence of condensation, for example, when supersaturation respect to ice occurs near the surface, and water vapor condenses onto the surface because of lack of ice nuclei.

Fig. 2.
Fig. 2.

(a) Example of SHI (below the 950-hPa level) caused by surface cooling and moisture condensation and (b) example of SHIs caused by humidity advection. The red lines are temperature, blue lines are dewpoint temperature, and thin dashed lines are specific humidity isolines. SHIs are the layers where the dewpoint temperature curves are tilted clockwise from the specific humidity isolines (dashed lines).

Citation: Journal of Climate 31, 10; 10.1175/JCLI-D-17-0497.1

Fig. 3.
Fig. 3.

Schematic illustration of processes leading or contributing to the formation of SHIs.

Citation: Journal of Climate 31, 10; 10.1175/JCLI-D-17-0497.1

Simultaneous occurrence of SHIs and TIs or SHIs and RHIs was utilized to evaluate the processes behind the formation of SHIs. Our hypothesis was that the simultaneous occurrence of SHIs and RHIs suggests that SHIs are formed as a result of humidity advection, whereas the simultaneous occurrence of SHIs and TIs in saturated conditions suggests that moisture condensation has impacted the formation of SHIs (Fig. 3). The criterion defining simultaneous occurrence was that inversions, either SHIs and TIs or SHIs and RHIs, should occur simultaneously, overlapping, in the layer examined. The simultaneous occurrence was examined for 50-hPa-thick layers. Hence, the probability of the simultaneous occurrence of SHIs with TIs or RHIs is not directly comparable to SHI occurrence. RHIs were computed from the relative humidity with respect to water (the choice of using RH with respect to water or with respect to water and ice only has a very small effect on the results), whereas the occurrence of SHIs in saturated conditions was computed from relative humidity with respect to water when the air temperature was above 0°C and with respect to ice when the air temperature was below 0°C. A relative humidity of 99%, instead of 100%, was used for the limit of saturated conditions, because of possible numerical inaccuracy. Condensation often occurs even before the gridcell mean relative humidity reaches 100%, but saturation in the entire gridcell volume is probably needed before the condensation substantially decreases the gridcell mean specific humidity. However, supersaturation with respect to ice is typical in the absence of ice nuclei. Therefore, it is not possible to set a universally valid relative humidity threshold for condensation.

The contribution of horizontal humidity advection to the strengthening of SHIs was examined by comparing the observed 6-h specific humidity tendencies against the tendencies calculated on the basis of horizontal specific humidity advection. Horizontal specific humidity advection was calculated by multiplying the specific humidity gradient along model-layer surfaces with the local horizontal wind speed. Linearity in the changes of humidity advection between analysis times was assumed. Only analysis times when the strength of SHIs increased between two consecutive model levels were used for the statistics. Based on these specific humidity tendencies, we calculated the contribution of differential humidity advection to the change of the specific humidity vertical gradient. For a further analysis, the contributions of humidity advection on SHI strength were averaged from layers between consecutive model levels to thicker layers. The computation of horizontal humidity advection from analysis fields with a low temporal resolution and over areas with substantial topographic variations is vulnerable to discretization errors (Seager and Henderson 2013). In addition, the assimilation of observations affects the tendencies in the reanalyses. Therefore, the advection calculations were not able to give quantitatively accurate results of the contribution of humidity advection to the strengthening of SHIs.

3. Results

a. Comparison of reanalyses and soundings

The accuracy of the reanalyses was evaluated by comparing them with the sounding data. Away from radiosonde sites, the reanalyses might be less accurate because of the lack of assimilation of radiosonde data and, particularly over sea ice and snow/ice-covered land, limited assimilation of infrared and microwave sounding data from satellites. However, radiosonde soundings may include errors. Typical errors are due to humidity sensor time lag, sensor icing, and sensor drying due to the heating of solar radiation (Anderson 1995; Miloshevich et al. 2006; Ingleby 2017). The time lag does not cause a bias in mean values but solar heating may cause a small dry bias, and sensor icing during an ascent through a cloud leads to a moist bias above the cloud (Miloshevich et al. 2006). Sensitivity to errors varies between radiosonde types (Miloshevich et al. 2006; Ingleby 2017). Hence, the differences between reanalyses and soundings do not always indicate the inaccuracy of reanalyses. The comparison at individual sounding sites was performed by choosing simultaneous data from soundings and from the closest grid point of the reanalyses. After that, the results were averaged for each region.

In both reanalyses, the mean profiles of specific humidity match the sounding data quite well (Figs. 4 and 5, left). However, the specific humidity in both reanalyses was typically lower than in the sounding data in winter (Fig. 4, center). JRA-55 was mostly drier than the soundings in all seasons with the largest differences occurring in summer over Siberia, where the difference exceeded 0.2 g kg−1 in the 500–700-hPa-layer (Fig. 5k). In ERA-Interim, the biases were smaller than in JRA-55, except in summer in the layer below 800 hPa (Fig. 5, center), where ERA-Interim was too moist. The summertime near-surface moist bias in ERA-Interim over the central Arctic Ocean has been reported in previous studies (Lüpkes et al. 2010; Jakobson et al. 2012; Wesslén et al. 2014). Our results indicated that the near-surface moist bias also occurs over continents. Another difference between ERA-Interim and the sounding data was in the midtroposphere over northern Europe and Siberia, where ERA-Interim was too dry. Over northern Europe and Siberia, both reanalyses had a dry bias in the 500–800-hPa layer. At least part of the dry bias in the reanalyses in comparison with the soundings can be explained by sensor icing problems in the radiosondes, notably those used in Russia (Ingleby 2017). On average, ERA-Interim had a smaller bias and a root-mean-squared error than JRA-55 in the comparison with radiosonde soundings.

Fig. 4.
Fig. 4.

(left) DJF mean specific humidity from ERA-Interim, JRA-55, and soundings. (center) Bias and (right) root-mean-square difference of ERA-Interim (dashed line) and JRA-55 (solid line) in the comparison with soundings in winter.

Citation: Journal of Climate 31, 10; 10.1175/JCLI-D-17-0497.1

Fig. 5.
Fig. 5.

As in Fig. 4, but for JJA.

Citation: Journal of Climate 31, 10; 10.1175/JCLI-D-17-0497.1

SHIs were on average stronger and more frequent in the sounding data than in the reanalyses (Fig. 6 and Table 1). Above the 900-hPa level, SHIs were more frequent in JRA-55 than ERA-Interim and vice versa below the 900-hPa level, but in the sounding data SHI occurrence was higher than in either of the reanalyses (Fig. 7). The difference in SHI occurrence between the reanalyses and the sounding data could be affected by their different vertical resolution. In the reanalyses, the vertical resolution decreases upward. Accordingly, at higher altitudes, reanalyses cannot resolve as many thin inversion layers as they can at lower altitudes, which may contribute to the upward-decreasing SHI occurrence in the reanalyses. The upward decrease in SHI occurrence was indeed larger in the reanalyses, especially in ERA-Interim, than in the soundings.

Fig. 6.
Fig. 6.

DJF and JJA means of SHI strength for the locations of the sounding stations from ERA-Interim (solid line), JRA-55 (long dashed line), and soundings (short dashed lines).

Citation: Journal of Climate 31, 10; 10.1175/JCLI-D-17-0497.1

Table 1.

Mean SHI occurrence and strength for regions of datasets. The last column, “Arctic,” covers the whole area north of 60°N. In each column, the first value is the mean value over whole region, and the second value is the mean value of the locations of the sounding stations only. SHI occurrence includes only the analyses and the grid points for which the surface pressure was higher than 800 hPa (for most of Greenland, the surface pressure is always smaller).

Table 1.
Fig. 7.
Fig. 7.

DJF and JJA means of SHI occurrence in 100-hPa-thick layers for the study regions from ERA-Interim and JRA-55 (solid lines) and for the locations of the sounding stations from ERA-Interim, JRA-55, and soundings (dashed lines).

Citation: Journal of Climate 31, 10; 10.1175/JCLI-D-17-0497.1

b. Specific humidity in the Arctic atmosphere

Specific humidity in the Arctic exhibits a large seasonal and spatial variation, linked to the variation of air temperature (Fig. 8). In winter, the mean vertically integrated water vapor content was lower than 10 kg m−2 almost everywhere north of 60°N, but in summer the values exceeded 10 kg m−2 everywhere, except over Greenland. The field of vertically integrated water vapor content was almost symmetric zonally in summer, except for the low values in Greenland, but in winter the lowest values occurred over three cold areas: Siberia, Greenland, and the Canadian archipelago. Relative humidity was usually high in the lower troposphere in winter, therefore air temperature limited the amount of water vapor in the lower troposphere, where most of atmospheric humidity is found (Fig. 8). Even though specific humidity typically increased downward, the layers where specific humidity decreased, that is, SHI layers, occurred frequently in the Arctic troposphere (Fig. 7). The occurrence of RHIs and TIs (Figs. 9 and 10), which occurred simultaneously with SHIs indicated that atmospheric conditions were different between the layers above and below approximately the 800-hPa level, so we present the properties of SHIs separately for the layers above and below the 800-hPa level.

Fig. 8.
Fig. 8.

(a) DJF and (b) JJA means of vertically integrated water vapor from ERA-Interim and soundings (circles). DJF and JJA means of specific humidity cross sections along latitudes (c),(d) 65°N and (e),(f) 75°N from ERA-Interim.

Citation: Journal of Climate 31, 10; 10.1175/JCLI-D-17-0497.1

Fig. 9.
Fig. 9.

DJF and JJA means of the fraction of SHIs that occurred simultaneously with TIs in 50-hPa-thick layers for the study regions from ERA-Interim and JRA-55 (solid lines) and for the locations of the sounding stations from ERA-Interim, JRA-55, and soundings (dashed lines).

Citation: Journal of Climate 31, 10; 10.1175/JCLI-D-17-0497.1

Fig. 10.
Fig. 10.

As in Fig. 9, but for SHIs that occurred simultaneously with RHIs.

Citation: Journal of Climate 31, 10; 10.1175/JCLI-D-17-0497.1

c. Specific humidity inversions above the 800-hPa level

SHIs were present less than 40% of the time in each 100-hPa-thick layer between the 400- and 800-hPa levels in both reanalyses, except in Greenland and in winter in Siberia and in the Alaska/Canada region, where SHIs were more frequent (Fig. 7). In the sounding data, the occurrence was higher, mostly between 40% and 60%. The difference between the reanalyses and the sounding data could not be explained by the locations of the sounding stations, because the occurrence profiles that were computed only from the grid points nearest to the sounding stations (dashed lines in Fig. 7) were representative for the whole region, but differed from those based on the sounding data. The seasonal and regional variations in SHI occurrence were smaller than the differences between the datasets.

Above the 800-hPa level, the formation of SHIs was mostly due to specific humidity advection, because almost all SHIs occurred simultaneously with RHIs but almost none with TIs in the 400–700-hPa layer in the reanalyses (Figs. 9 and 10). In the reanalyses, TIs were rare in this layer (occurrence mostly below 1% in each 100-hPa-thick layer). Mean relative humidity was mostly below 50% in the 400–800-hPa layer when SHIs occurred in both reanalyses. In the sounding data, the proportion of the SHIs occurring simultaneously with RHIs was smaller (Fig. 10), and the proportion of the SHIs occurring simultaneously with TIs was larger than in the reanalyses (Fig. 9).

d. Specific humidity inversions below the 800-hPa level

SHI occurrence below the 800-hPa level was dependent on the surface heat budget. In winter SHI occurrence was high over the continents and ice-covered seas (Figs. 11a,b), where the negative net longwave radiation effectively cooled the surface. In summer SHI occurrence in the reanalyses was highest over the Arctic Ocean (Figs. 11c,d), where sea ice and snowmelt and the large heat capacity of the open sea limited the seasonal increase of the near-surface air temperature. Spring and autumn were transition periods between winter and summer types of SHIs (Fig. 12). In addition, over sloping surfaces, especially over slopes near the coasts of Greenland, SHIs associated with RHIs and formed via differential humidity advection (Fig. 13) were identified. The formation mechanism of these SHIs was probably a consequence of dry and cold air advection from an ice sheet to a near-surface coastal layer as suggested for Antarctica by Nygård et al. (2013).

Fig. 11.
Fig. 11.

(a),(b) DJF and (c),(d) JJA means of SHI occurrence between the level of 800 hPa and the surface from ERA-Interim, JRA-55, and soundings (circles).

Citation: Journal of Climate 31, 10; 10.1175/JCLI-D-17-0497.1

Fig. 12.
Fig. 12.

Monthly means of SHI strength from ERA-Interim.

Citation: Journal of Climate 31, 10; 10.1175/JCLI-D-17-0497.1

fig. 13.
fig. 13.

The contribution of specific humidity advection to the increase of SHI strength on layers between subsequent model levels below the 800-hPa level based on ERA-Interim.

Citation: Journal of Climate 31, 10; 10.1175/JCLI-D-17-0497.1

1) Specific humidity inversions in winter

The formation of winter SHIs began in October over the northernmost land areas and ice-covered seas (Fig. 12). The strength and the areal extent of SHIs increased during November and December with decreasing solar radiation and decreased in spring with increasing solar radiation (Fig. 12.). In winter, SHI occurrence below the 800-hPa level was high, exceeding 90%, over cold surfaces over continents and the Arctic Ocean, and low, less than 50%, over the warm sea surface of the North Atlantic in both reanalyses (Fig. 11 and Table 1). Vertically, SHIs were most common in the layer below 900 hPa (Fig. 7). The differences between the reanalyses and between the reanalyses and the sounding data were small in the areas of high SHI occurrence. In JRA-55, SHI occurrence was slightly higher over sea ice than in ERA-Interim (Fig. 11). In Greenland, SHI occurrence could not be properly presented at constant pressure levels because of the orography. In Figs. 7, 9, and 10 the curves below the 800-hPa level only represent the coastal areas of Greenland, because the surface pressure is always lower than 800 hPa in most of Greenland (masked area in Fig. 11). Nevertheless, SHIs between the two lowest model levels were most frequent in Greenland.

On continents and the Arctic Ocean, most wintertime SHIs below the 800-hPa level occurred simultaneously with TIs (Fig. 9), and the occurrence with RHIs was smaller than at higher altitudes (Fig. 10). At low levels, SHIs normally occurred in conditions of high relative humidity, and saturated conditions (relative humidity over 99%) were present on average in 10%–30% of the cases when an SHI occurred in the reanalyses. Accordingly, SHIs occurred in conditions when condensation was able to form or strengthen SHIs. The importance of the contribution of condensation in the formation of SHIs was supported by the result that in most continental areas and the Atlantic Ocean, the contribution of vertically differential horizontal humidity advection was less than 50% (Fig. 13).

The mean SHI strength was lower in the reanalyses, especially in JRA-55, than in the sounding data (Fig. 6). The highest mean SHI strength occurred inland in northern Canada and Alaska (Fig. 14) in all datasets, but the differences in the spatial distribution of the mean SHI strength between the reanalyses were much larger than in SHI occurrence. In ERA-Interim, the highest values of the mean SHI strength occurred over continents and near the eastern coast of Greenland. In addition to the areas of strong SHIs in northern Canada and Alaska and the eastern coast of Greenland, high values of the mean SHI strength were also found over sea ice near open seas near the Bering Strait and north of Svalbard in JRA-55. On continents, the mean SHI strength was higher in ERA-Interim than in JRA-55 because of a larger vertical gradient of specific humidity in SHIs. On the contrary, over sea ice, the mean SHI strength in JRA-55 was higher than in ERA-Interim. In ERA-Interim, the mean SHI strength was smaller over sea ice than over land, but in JRA-55, there was no difference in the mean SHI strength between these surface types. Over the Arctic sea ice, the latent heat flux was upward in ERA-Interim, whereas in JRA-55 it was downward, which probably at least partly explains the moister conditions at the lowest model level and weaker SHIs over sea ice in ERA-Interim than in JRA-55. However, the magnitudes of the latent heat fluxes were small in both reanalyses.

fig. 14.
fig. 14.

(a),(b) DJF and (c),(d) JJA means of SHI strength from ERA-Interim, JRA-55, and soundings (circles).

Citation: Journal of Climate 31, 10; 10.1175/JCLI-D-17-0497.1

2) Specific humidity inversions in summer

The strongest SHIs in summer formed over seas when warm, moist air was advected from land to over a relatively cold sea surface. Even though SHIs were also present over continents, they were fewer (Fig. 11) and weaker (Fig. 14) than SHIs over cold seas. The season of summer-type SHIs over the Arctic Ocean began in June, when the air specific humidity over continents exceeded the saturation specific humidity of the sea surface, and ended in September (Fig. 12).

An example of the formation of summer-type SHIs is presented in Fig. 15. The formation of SHIs over the East Siberian Sea and the Arctic Ocean on 6–8 August 2014 was caused by the offshore flow of moist air from Siberia. The specific humidity maximum occurred below the 900-hPa level near the Siberian coast, but its height increased with the fetch over the sea, as did the height of the air temperature maximum; these increases were due to upward-increasing moisture and heat transport and the cooling and drying of the lowest layers of the advected air mass. The wind direction was almost the same in the entire layer of the major humidity advection from the surface up to the 700-hPa level, and wind speed increased only slightly upward from the 900- to 750-hPa level. The maximum moisture transport occurred between the 850- and 900-hPa levels, almost collocating with the layer of the maximum specific humidity. Relative humidity was at the saturation point below the humidity maximum, and both sensible and latent heat fluxes were downward below the SHI, resulting in cooling and drying, which was also contributed to by the condensation of moisture of the advected air mass. These processes also decreased specific humidity at the level of the humidity maximum, causing a weakening of SHI with increasing distance from the coast.

fig. 15.
fig. 15.

(a),(c),(e) SHI strength and moisture flux density at the level of 850 hPa (black arrows) from ERA-Interim. (b),(d),(f) Specific humidity cross section along longitude 150°E [marked with a red line in (a), (c), and (e)] and meridional moisture flux density (black arrows) from ERA-Interim.

Citation: Journal of Climate 31, 10; 10.1175/JCLI-D-17-0497.1

At the same time as the example case, the Swedish icebreaker Oden was in the East Siberian Sea slightly eastward of the location of the example case, where almost similar offshore humidity advection occurred a couple of days later. The ERA-Interim values for the maximum specific humidity, over 10 g kg−1, and the altitude of the humidity maximum near the coast in the example case (Fig. 15) were close to the values observed at Oden by Tjernström et al. (2015), but the surface fluxes of sensible heat and latent heat had a much larger magnitude than the measured fluxes reported by Tjernström et al. (2015).

In summer, SHIs were most frequent over the Arctic Ocean (Table 1). The maximum occurrence was in both reanalyses near the coast of the Arctic Ocean, and the occurrence was higher in ERA-Interim than JRA-55 (Fig. 11). Over the Arctic Ocean, the difference between the reanalyses increased with distance from the coast, so in ERA-Interim SHIs frequently penetrated farther over the sea. In summer, near-surface SHIs (SHIs between the two lowest model levels) were frequent in central Greenland but they occurred above the 800-hPa level.

The mean SHI strength was largest near the coast of the Arctic Ocean and decreased toward the North Pole and the northern coast of Greenland (Fig. 14). Summer SHIs near the coast of the Arctic Ocean were even stronger than winter SHIs over continents (Fig. 14). Stronger SHIs in summer than in winter have been also reported in previous studies (Devasthale et al. 2011; Nygård et al. 2014; Brunke et al. 2015). In JRA-55, the mean SHI strength over the Arctic Ocean was only 40% of that in ERA-Interim (Table 1). In the sounding data, SHIs were weak in summer, and the differences between the sounding data and the reanalyses were smaller in summer than in winter (Fig. 6). However, the strongest SHIs were not well represented in the sounding data, because the sounding stations are located on land. As strong SHIs over the sea occurred during offshore winds, the sounding stations were in these cases under an influence of continental conditions. However, when strong SHIs occurred over land near the coast of the Arctic Ocean, especially in Siberia, the near-surface winds were directed from sea to inland areas.

3) Specific humidity inversions generated by downslope winds

Over a sloping surface, especially in the coastal zone of Greenland, SHIs were often generated by the advection of dry air due to katabatic or other downslope winds. For these areas, the effect of advection dominated the formation of SHIs (Fig. 13). In these conditions, the relative humidity at the lowest model level was even below 70%, and the relative humidity was lower at the lowest model than at the level of the specific humidity maximum, but the temperature difference between those levels was small. Over the slopes of the Greenland ice sheet, high simultaneous occurrences of SHIs and RHIs were found from the approximately 30-hPa level above the surface when SHI occurrence was examined on model levels. Accordingly, an RHI rather than a TI explained the formation of SHIs over the slopes of Greenland. The formation of a dry-air layer requires the formation of a cold, dry air mass on a highly elevated area and the downslope advection of this air mass. When adiabatic warming decreased the relative humidity faster than evaporation increased it, a layer of low relative humidity formed near the surface, which could also be seen in the reanalyses’ surface fluxes: a strong downward sensible heat flux was present in these areas, but the latent heat flux was upward.

4. Discussion

We examined the spatial distribution of SHI occurrence based on reanalyses and sounding data. In general, SHI occurrence was as high as in previous studies (Brunke et al. 2015; Nygård et al. 2014; Devasthale et al. 2011). We found that the seasonal variation of SHI occurrence was large over the continents but smaller over the oceans, and seasonal and spatial variations occurred mostly below the 800-hPa level. The amount of spatial variation in SHI occurrence depended on the thickness of the examined layer, in such a way that, for large areas in winter, the occurrence was close to 100% in the layer below the 800-hPa level (Fig. 11). The spatial variation of SHI occurrence became smaller when a thicker layer was examined. The above-mentioned effect is a consequence of the simultaneous occurrence of SHIs in several layers. Nygård et al. (2014), addressing the layer up to the 500-hPa level, reported seasonal and spatial variations smaller than found here in the layer up to the 800-hPa level. They reported that in the layer up to the 500-hPa level in every season, SHI occurrence was over 70% at all Arctic sounding stations, except Turuhansk in Siberia. Our findings based on the sounding data indicated that SHI occurrence below the 800-hPa level was below 70% at almost half of the sounding stations in summer and at three stations (near the Atlantic Ocean) in winter. In winter, the spatial distribution of SHI occurrence below the 800-hPa level was rather similar to that presented by Devasthale et al. (2011) based on AIRS data for the layer between the surface and the 400-hPa level. In summer over the Arctic Ocean, our results indicated a much higher SHI occurrence than the results of Devasthale et al. (2011). At least a part of the difference could be explained by the fact that their data did not well represent summer conditions over the Arctic Ocean, because they could only use clear-sky observations, but cloudy conditions prevail over the Arctic Ocean (Intrieri et al. 2002b).

Nygård et al. (2014) and Brunke et al. (2015) have discussed the role of advection and moisture condensation for the formation of SHIs. We found that atmospheric conditions related to SHIs were notably different below and above the 800-hPa level. Above the 800-hPa level during the occurrence of SHIs, the stratification was weaker and relative humidity was lower than in the layer below 800 hPa, which allowed upward-increasing heat and humidity advection and the formation of SHIs without the formation of TIs. Above the 800-hPa level, SHIs occurred at the boundaries between dry and moist air masses, which were probably formed when a moist air mass was advected above a dry air mass. The computation of the effect of differential humidity advection on strengthening SHIs showed that differential humidity advection explained only approximately 50% of the increase in SHI strength among all strengthening SHIs above the 800-hPa level. However, the method to evaluate the evolution of SHI strength from advection profiles did not take into account the effect of data assimilation, which could affect this result.

In the boundary layer, the cooling of an air mass over a cold surface leads to high relative humidity and very stable stratification and then to the occurrence of TIs and SHIs. Our results clearly indicated that the frequent occurrence of near-surface SHIs in the Arctic is closely connected to 1) the surface cooling caused by the negative net radiation in winter and 2) the cold surfaces of both ice-covered and open parts of the Arctic Ocean in summer (Fig. 11). This suggests that moisture condensation related to near-surface air mass cooling is an important factor for generating SHIs below 800 hPa, which is supported by the fact that most of the SHIs below the 800-hPa level occurred simultaneously with TIs (Fig. 9) and high relative humidity. In winter, the formation of TIs and SHIs is conventionally thought to be due to radiative surface cooling in the conditions of clear skies or optically thin clouds, which is often the case in the Artic (Stramler et al. 2011), but SHIs are also present in cloudy conditions (Stramler et al. 2011; Sedlar et al. 2012; Nygård et al. 2014; Sedlar 2014). Clouds modify radiative energy transfer, increasing the downward longwave radiation at the surface (Shupe and Intrieri 2004). In winter, cloudy conditions (Stramler et al. 2011) and increased humidity advection (Woods et al. 2013) are often associated with synoptic-scale cyclones, and both of them probably affect the specific humidity profile. Above the 800-hPa level, a small part of SHIs occurred without the presence of a TI (Fig. 9), and the majority of SHIs (Fig. 10) occurred simultaneously with an RHI, suggesting the effects of humidity advection on SHI occurrence (process 1 in Fig. 3). Nygård et al. (2013, 2014) proposed that an upward-increasing horizontal moisture transport–related large-scale moisture convergence may be important for the formation of SHIs, and Brunke et al. (2015) showed that the effect of differential humidity advection can strengthen or weaken SHIs depending on the location. Our results also indicate that the differential humidity advection contributed to the strengthening of SHIs but did not entirely explain the formation of SHIs (Fig. 13). Devasthale et al. (2011) suggested that a TI prevents effective moisture transport near the surface and the downward turbulent mixing of moisture from the layer of the maximum horizontal moisture transport, and in that way, affects the formation of SHIs (process 5 in Fig. 3). Accordingly, humidity advection may contribute to the occurrence of an SHI or also increase its strength in cases when the SHI was originally generated via surface cooling and condensation (process 4 in Fig. 3).

The main formation mechanism of summer SHIs is the advection of warm, moist air over a cold surface, causing condensation and the removal of moisture from the lowest parts of the boundary layer. An important difference between summer and winter is that in summer the Arctic Ocean surface temperature typically remains between −2°C and 0°C (Lüpkes et al. 2010; Tjernström et al. 2012; Sotiropoulou et al. 2016) because of the large heat capacity of the ocean and the presence of (melting) snow or ice. Therefore, in summer SHI strength is related to the specific humidity of the advected air mass, which agrees with the suggestion of Brunke et al. (2015) that regional moisture convergence in the Arctic is related to the mean strength of SHIs. Instead, in winter, the surface cooling due to a negative net radiation may result in the deepening of TIs and SHIs without heat or humidity advection. In summer over the Arctic Ocean, air mass cooling and moisture condensation at the level of the specific humidity maximum result in weakening SHIs with increasing fetch over the cold sea, which reduces the area of the occurrence of strong SHIs over a cold sea surface. Previous studies (Solomon et al. 2011, 2014; Sedlar et al. 2012; Shupe et al. 2013) have shown that an SHI layer above a cloud layer can be a moisture source for the cloud layer. Thus, turbulent downward moisture transport from the SHI layer to the cloud layer together with moisture condensation in clouds or in fog are probably important for the drying of the advected air mass. This is supported by our result that the layer of near-surface cooling and drying grew higher with increasing distance from the coast (Fig. 15), as well as by previous observations of the frequent occurrence of 100% relative humidity (Tjernström et al. 2012; Sotiropoulou et al. 2016), fog, and extensive low cloudiness (Intrieri et al. 2002b; Tjernström 2005; Sotiropoulou et al. 2016), which indicate that moisture condensation is common over the Arctic Ocean. Upward motions with adiabatic cooling and moisture condensation can also contribute to air mass drying.

Based on the comparison with radiosonde soundings, the reanalyses are able to represent the main features of the spatial variation (between sounding sites) of the occurrence and strength of SHIs. However, the mean SHI strength was underestimated in both reanalyses. In winter, the error was notably larger for JRA-55 than ERA-Interim, and this was due to a smaller vertical specific humidity gradient in JRA-55. This suggests that vertical mixing in JRA-55 is probably too strong in very stable conditions, which prevents the formation of strong inversions. The stronger vertical mixing in JRA-55 can also explain the moister conditions at the lowest model level over continents and the differences in SHI properties over the Arctic Ocean in summer: weaker vertical mixing in ERA-Interim allowed moist air advection to penetrate farther to the Arctic Ocean without strong cooling and drying by turbulent heat fluxes.

Soundings are mostly performed twice a day, whereas the time resolution of the reanalysis products consisted of four analyses a day. This probably only caused a minimal effect on the comparison of the reanalyses and soundings, because the diurnal cycle of specific humidity was weak. In the reanalyses, the largest amplitude of the diurnal cycle was observed over continents in summer, where it possibly affected SHI occurrence near the surface. ERA-Interim showed that the near-surface specific humidity had two maximums and minimums a day. The maximums occurred between 0600–1200 and 1800–2400 local solar time, and the minimums occurred between 0000–0600 and 1200–1800 local solar time. The nocturnal minimum was related to the SHI caused by surface radiative cooling. If soundings had not been performed in nighttime (0000–0600 local solar time), this SHI could not be detected.

SHI occurrence, and especially SHI strength, was sensitive to surface properties and especially surface heat fluxes. Upward latent heat fluxes from ice-covered seas to the atmosphere in ERA-Interim compared with downward latent heat fluxes in JRA-55 were probably at least partly responsible for the weaker SHIs over sea ice in ERA-Interim than in JRA-55. For the summer cruise examined by Wesslén et al. (2014), the upward latent heat flux was overestimated in ERA-Interim over the central Arctic Ocean, and the downward sensible and latent heat fluxes in our example case were larger than the fluxes measured by Tjernström et al. (2015) during a similar episode as the example case. Further, Tastula et al. (2013) showed that reanalysis products for the turbulent surface fluxes of sensible and latent heat had very weak or even negative correlations with in situ observations over Antarctic sea ice. Differences in surface heat fluxes between the observations and reanalyses caused uncertainty in the reanalysis-based estimates of SHI properties.

5. Conclusions

For the first time, we showed that the occurrence and properties of SHIs and their formation mechanisms vary with altitude. Arctic SHIs can be divided into two main categories based on their physical properties: 1) SHIs below the 800-hPa level and 2) SHIs above the 800-hPa level. The key properties of SHIs can be summarized as follows:

  1. Above the 800-hPa level, SHIs occurred with RHIs and without TIs and in conditions of low relative humidity, which suggests that they were formed when a moist air mass was advected over a dry air mass. SHI occurrence above the 800-hPa level had relatively small spatial and seasonal variations.
  2. SHI occurrence below the 800-hPa level had large seasonal and spatial variations, which depended on the surface heat budget, and SHIs occurred with TIs in conditions of high relative humidity.
  3. SHIs below the 800-hPa level can be divided in three groups: (i) SHIs in winter, formed because of radiative cooling, (ii) SHIs in summer generated by warm, moist air advection over a cold sea surface, and (iii) SHIs generated by differential humidity advection due to downslope winds.
  4. In winter, SHI occurrence is high over continents and the Arctic Ocean and low over the Atlantic Ocean. In summer, SHI occurrence is high over the cold sea surface of the Arctic Ocean and low over continents and the Atlantic Ocean.
  5. SHIs were strongest over the Arctic Ocean in summer because of the larger water vapor content of air in summer than in winter, which allows for larger vertical specific humidity gradients.
  6. SHIs generated by downslope winds occurred throughout the year mostly in coastal areas of Greenland and were formed when cold and dry air from the ice sheet was advected in a shallow near-surface layer.
The comparison to radiosonde soundings showed that reanalyses were able to describe the mean vertical profiles of specific humidity and the seasonal and spatial variations of SHI occurrence and strength. However, SHI occurrence above the 800-hPa level and SHI strength were underestimated in the reanalyses, the latter particularly in JRA-55. The vertical resolution decreases upward in the reanalyses, so that very thin SHI layers cannot be resolved, which generated differences between the reanalyses and the sounding data. Uncertainty in turbulent surface fluxes and vertical mixing probably caused differences in SHI strength between the reanalyses and the sounding data.

These results provide a more detailed view of the processes involved in SHI formation. To better understand the Arctic climate system, further studies on the interactions between SHIs and the vertical distributions of clouds and moisture transport to the Arctic, as well as the associated radiative and turbulent processes, are needed.

Acknowledgments

The work was supported by the Academy of Finland via the TWASE Project (Contract 283101) and the INTAROS Project (Grant 727890) funded by the Horizon 2020 Programme of the European Commission. We thank the ECMWF for providing us with the ERA-Interim reanalysis and the Japan Meteorological Agency for the JRA-55 reanalysis.

REFERENCES

  • Anderson, P. S., 1995: Mechanism for the behavior of hydroactive materials used in humidity sensors. J. Atmos. Oceanic Technol., 12, 662667, https://doi.org/10.1175/1520-0426(1995)012<0662:MFTBOH>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Andreas, E. L., P. S. Guest, P. O. G. Persson, C. W. Fairall, T. W. Horst, R. E. Moritz, and S. R. Semmer, 2002: Near-surface water vapor over polar sea ice is always near ice saturation. J. Geophys. Res., 107, 8033, https://doi.org/10.1029/2000JC000411.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Boisvert, L. N., and J. C. Stroeve, 2015: The Arctic is becoming warmer and wetter as revealed by the Atmospheric Infrared Sounder. Geophys. Res. Lett., 42, 44394446, https://doi.org/10.1002/2015GL063775.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bring, A., and Coauthors, 2016: Arctic terrestrial hydrology: A synthesis of processes, regional effects, and research challenges. J. Geophys. Res. Biogeosci., 121, 621649, https://doi.org/10.1002/2015JG003131.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Brunke, M. A., S. T. Stegall, and X. Zeng, 2015: A climatology of tropospheric humidity inversions in five reanalyses. Atmos. Res., 153, 165187, https://doi.org/10.1016/j.atmosres.2014.08.005.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Curry, J., 1983: On the formation of continental polar air. J. Atmos. Sci., 40, 22782292, https://doi.org/10.1175/1520-0469(1983)040<2278:OTFOCP>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dee, D. P., and Coauthors, 2011: The ERA‐Interim reanalysis: Configuration and performance of the data assimilation system. Quart. J. Roy. Meteor. Soc., 137, 553597, https://doi.org/10.1002/qj.828.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Devasthale, A., J. Sedlar, and M. Tjernström, 2011: Characteristics of water-vapour inversions observed over the Arctic by Atmospheric Infrared Sounder (AIRS) and radiosondes. Atmos. Chem. Phys., 11, 98139823, https://doi.org/10.5194/acp-11-9813-2011.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dufour, A., O. Zolina, and S. K. Gulev, 2016: Atmospheric moisture transport to the Arctic: Assessment of reanalyses and analysis of transport components. J. Climate, 29, 50615081, https://doi.org/10.1175/JCLI-D-15-0559.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Durre, I., and X. Yin, 2008: Enhanced radiosonde data for studies of vertical structure. Bull. Amer. Meteor. Soc., 89, 12571262.

  • Durre, I., R. S. Vose, and D. B. Wuertz, 2006: Overview of the integrated global radiosonde archive. J. Climate, 19, 5368, https://doi.org/10.1175/JCLI3594.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Francis, J. A., D. M. White, J. J. Cassano, W. J. Gutowski Jr., L. D. Hinzman, M. M. Holland, M. A. Steele, and C. J. Vörösmarty, 2009: An Arctic hydrologic system in transition: Feedbacks and impacts on terrestrial, marine, and human life. J. Geophys. Res., 114, G04019, https://doi.org/10.1029/2008JG000902.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gettelman, A., V. P. Walden, L. M. Miloshevich, W. L. Roth, and B. Halter, 2006: Relative humidity over Antarctica from radiosondes, satellites, and a general circulation model. J. Geophys. Res., 111, D09S13, https://doi.org/10.1029/2005JD006636.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ingleby, B., 2017: An assessment of different radiosonde types 2015/2016. ECMWF Tech. Memo. 807, 69 pp., https://www.ecmwf.int/sites/default/files/elibrary/2017/17551-assessment-different-radiosonde-types-20152016.pdf.

  • Intrieri, J. M., C. W. Fairall, M. D. Shupe, P. O. G. Persson, E. L. Andreas, P. S. Guest, and R. E. Moritz, 2002a: An annual cycle of Arctic surface cloud forcing at SHEBA. J. Geophys. Res., 107, 8039, https://doi.org/10.1029/2000JC000439.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Intrieri, J. M., M. D. Shupe, T. Uttal, and B. J. McCarty, 2002b: An annual cycle of Arctic cloud characteristics observed by radar and lidar at SHEBA. J. Geophys. Res., 107, 8030, https://doi.org/10.1029/2000JC000423.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Jakobson, E., and T. Vihma, 2010: Atmospheric moisture budget in the Arctic based on the ERA-40 reanalysis. Int. J. Climatol., 30, 21752194, https://doi.org/10.1002/joc.2039.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Jakobson, E., T. Vihma, T. Palo, L. Jakobson, H. Keernik, and J. Jaagus, 2012: Validation of atmospheric reanalyses over the central Arctic Ocean. Geophys. Res. Lett., 39, L10802, https://doi.org/10.1029/2012GL051591.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kobayashi, S., and Coauthors, 2015: The JRA-55 reanalysis: General specifications and basic characteristics. J. Meteor. Soc. Japan, 93, 548, https://doi.org/10.2151/jmsj.2015-001.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lüpkes, C., T. Vihma, E. Jakobson, G. König‐Langlo, and A. Tetzlaff, 2010: Meteorological observations from ship cruises during summer to the central Arctic: A comparison with reanalysis data. Geophys. Res. Lett., 37, L09810, https://doi.org/10.1029/2010GL042724.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Miloshevich, L. M., H. Vömel, D. N. Whiteman, B. M. Lesht, F. J. Schmidlin, and F. Russo, 2006: Absolute accuracy of water vapor measurements from six operational radiosonde types launched during AWEX‐G and implications for AIRS validation. J. Geophys. Res., 111, D09S10, https://doi.org/10.1029/2005JD006083.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Nygård, T., T. Valkonen, and T. Vihma, 2013: Antarctic low-tropospheric humidity inversions: 10-yr climatology. J. Climate, 26, 52055219, https://doi.org/10.1175/JCLI-D-12-00446.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Nygård, T., T. Valkonen, and T. Vihma, 2014: Characteristics of Arctic low-tropospheric humidity inversions based on radio soundings. Atmos. Chem. Phys., 14, 19591971, https://doi.org/10.5194/acp-14-1959-2014.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Persson, P. O. G., C. W. Fairall, E. L. Andreas, P. S. Guest, and D. K. Perovich, 2002: Measurements near the Atmospheric Surface Flux Group tower at SHEBA: Near‐surface conditions and surface energy budget. J. Geophys. Res., 107, 8045, https://doi.org/10.1029/2000JC000705.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Prata, A. J., 1996: A new long-wave formula for estimating downward clear-sky radiation at the surface. Quart. J. Roy. Meteor. Soc., 122, 11271151, https://doi.org/10.1002/qj.49712253306.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Savré, J., A. M. L. Ekman, G. Svensson, and M. Tjernström, 2015: Large-eddy simulations of an Arctic mixed-phase stratiform cloud observed during ISDAC: Sensitivity to moisture aloft, surface fluxes and large-scale forcing. Quart. J. Roy. Meteor. Soc., 141, 11771190, https://doi.org/10.1002/qj.2425.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Seager, R., and N. Henderson, 2013: Diagnostic computation of moisture budgets in the ERA-Interim reanalysis with reference to analysis of CMIP-archived atmospheric model data. J. Climate, 26, 78767901, https://doi.org/10.1175/JCLI-D-13-00018.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sedlar, J., 2014: Implications of limited liquid water path on static mixing within Arctic low-level clouds. J. Appl. Meteor. Climatol., 53, 27752789, https://doi.org/10.1175/JAMC-D-14-0065.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sedlar, J., M. D. Shupe, and M. Tjernström, 2012: On the relationship between thermodynamic structure and cloud top, and its climate significance in the Arctic. J. Climate, 25, 23742393, https://doi.org/10.1175/JCLI-D-11-00186.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Serreze, M. C., and Coauthors, 2006: The large-scale freshwater cycle of the Arctic. J. Geophys. Res., 111, C11010, https://doi.org/10.1029/2005JC003424.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Shupe, M. D., and J. M. Intrieri, 2004: Cloud radiative forcing of the Arctic surface: The influence of cloud properties, surface albedo, and solar zenith angle. J. Climate, 17, 616628, https://doi.org/10.1175/1520-0442(2004)017<0616:CRFOTA>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Shupe, M. D., P. O. G. Persson, I. M. Brooks, M. Tjernström, J. Sedlar, T. Mauritsen, S. Sjogren, and C. Leck, 2013: Cloud and boundary layer interactions over the Arctic sea ice in late summer. Atmos. Chem. Phys., 13, 93799400, https://doi.org/10.5194/acp-13-9379-2013.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Solomon, A., M. D. Shupe, P. O. G. Persson, and H. Morrison, 2011: Moisture and dynamical interactions maintaining decoupled Arctic mixed-phase stratocumulus in the presence of a humidity inversion. Atmos. Chem. Phys., 11, 10 12710 148, https://doi.org/10.5194/acp-11-10127-2011.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Solomon, A., M. D. Shupe, P. O. G. Persson, H. Morrison, T. Yamaguchi, P. M. Caldwell, and G. de Boer, 2014: The sensitivity of springtime Arctic mixed-phase stratocumulus clouds to surface-layer and cloud-top inversion-layer moisture sources. J. Atmos. Sci., 71, 574595, https://doi.org/10.1175/JAS-D-13-0179.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sotiropoulou, G., and Coauthors, 2016: Atmospheric conditions during the Arctic Clouds in Summer Experiment (ACSE): Contrasting open water and sea ice surfaces during melt and freeze-up seasons. J. Climate, 29, 87218744, https://doi.org/10.1175/JCLI-D-16-0211.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Stramler, K., A. D. Del Genio, and W. B. Rossow, 2011: Synoptically driven Arctic winter states. J. Climate, 24, 17471762, https://doi.org/10.1175/2010JCLI3817.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Susskind, J., J. M. Blaisdell, and L. Iredell, 2014: Improved methodology for surface and atmospheric soundings, error estimates, and quality control procedures: The Atmospheric Infrared Sounder science team version-6 retrieval algorithm. J. Appl. Remote Sens., 8, 084994, https://doi.org/10.1117/1.JRS.8.084994.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Tastula, E.-M., T. Vihma, E. L. Andreas, and B. Galperin, 2013: Validation of the diurnal cycles in atmospheric reanalyses over Antarctic sea ice. J. Geophys. Res. Atmos., 118, 41944204, https://doi.org/10.1002/jgrd.50336.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Tjernström, M., 2005: The summer Arctic boundary layer during the Arctic Ocean Experiment 2001 (AOE-2001). Bound.-Layer Meteor., 117, 536, https://doi.org/10.1007/s10546-004-5641-8.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Tjernström, M., and Coauthors, 2012: Meteorological conditions in the central Arctic summer during the Arctic Summer Cloud Ocean Study (ASCOS). Atmos. Chem. Phys., 12, 68636889, https://doi.org/10.5194/acp-12-6863-2012.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Tjernström, M., and Coauthors, 2015: Warm‐air advection, air mass transformation and fog causes rapid ice melt. Geophys. Res. Lett., 42, 55945602, https://doi.org/10.1002/2015GL064373.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Vihma, T., T. Kilpeläinen, M. Manninen, A. Sjöblom, E. Jakobson, T. Palo, J. Jaagus, and M. Maturilli, 2011: Characteristics of temperature and humidity inversions and low-level jets over Svalbard fjords in spring. Adv. Meteor., 2011, 486807, https://doi.org/10.1155/2011/486807.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Vihma, T., and Coauthors, 2016: The atmospheric role in the Arctic water cycle: A review on processes, past and future changes, and their impacts. J. Geophys. Res. Biogeosci., 121, 586620, https://doi.org/10.1002/2015JG003132.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wesslén, C., M. Tjernström, D. H. Bromwich, G. de Boer, A. M. L. Ekman, L.-S. Bai, and S.-H. Wang, 2014: The Arctic summer atmosphere: An evaluation of reanalyses using ASCOS data. Atmos. Chem. Phys., 14, 26052624, https://doi.org/10.5194/acp-14-2605-2014.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Woods, C., R. Caballero, and G. Svensson, 2013: Large‐scale circulation associated with moisture intrusions into the Arctic during winter. Geophys. Res. Lett., 40, 47174721, https://doi.org/10.1002/grl.50912.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, T., K. Stamnes, and S. A. Bowling, 2001: Impact of the atmospheric thickness on the atmospheric downwelling longwave radiation and snowmelt under clear-sky conditions in the Arctic and Subarctic. J. Climate, 14, 920939, https://doi.org/10.1175/1520-0442(2001)014<0920:IOTATO>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
Save