Surface Energy Balance Framework for Arctic Amplification of Climate Change [Journal of Climate]
(Journal of Climate Via Acquire Media NewsEdge) ABSTRACT
Using 22 Canadian radiosonde stations from 1971 to 2010, the annually averaged surface air temperature trend amplification ranged from 1.4 to 5.2 relative to the global average warming of 0.178C decade-1. The amplification factors exhibit a strong latitudinal dependence varying from 2.6 to 5.2 as the latitude increases from 508 to 808N. The warming trend has a strong seasonal dependence with the greatest warming taking place from September to April. The monthly variations in the warming trend are shown to be related to the surface-based temperature inversion strength and the mean monthly surface air temperatures.
The surface energy balance (SEB) equation is used to relate the response of the surface temperature to changes in the surface energy fluxes. Based on the SEB analysis, there are four contributing factors to Arctic amplification: 1) a larger change in net downward radiation at the Arctic surface compared to the global average; 2) a larger snow and soil conductive heat flux change than the global average; 3) weaker sensible and latent heat flux responses that result in a larger surface temperature response in the Arctic; and 4) a colder skin temperature compared to the global average, which forces a larger surface warming to achieve the same increase in upward longwave radiation. The observed relationships between the Canadian station warming trends and both the surface-based inversion strength and the surface air temperature are shown to be consistent with the SEB analysis. Measurements of conductive flux were not available at these stations.
(ProQuest: ... denotes formulae omitted.)
The enhanced surface warming in the Arctic compared to the global average during the last several decades is well documented (Hansen et al. 2006; Solomon et al. 2007). The phenomenon, known as Arctic or polar amplification (e.g., Polyakov et al. 2002; Serreze and Francis 2006), implies that the Arctic surface temperature is inherently more sensitive to both natural and anthropogenic forcings. This bestows on the Arctic environment a special role for the detection of the influence of anthropogenic climate change.
Anumber of processes have been shown to contribute to Arctic amplification: ice/snow albedo reduction (e.g., Manabe and Stouffer 1980), sea ice volume reduction accompanied by a larger ocean to air heat flux (Comiso et al. 2008; Serreze et al. 2009; Screen and Simmonds 2010b,a), enhanced meridional energy transport from the middle and tropical latitudes (Alexeev et al. 2005; Graversen and Wang 2009; Yang et al. 2010), cloud shortwave albedo and longwave emissivity changes (Garrett et al. 2002; Garrett and Zhao 2006; Liu et al. 2008), and changes in atmospheric aerosol properties (Shindell and Faluvegi 2009; Mauritsen et al. 2011). Global climate models have been used to examine how various global feedbacks affect model predicted amplification (e.g., Holland and Bitz 2003; Lu and Cai 2010). Serreze and Barry (2011) summarize the various mechanisms in a recent review article and conclude that multiple causes, as listed above, are operating but that the strong warming over the Arctic Ocean during the past decade in autumn and winter is associated with sea ice loss.
The present paper examines how surface energy balance (SEB), the energy constraint given by the sum of all the energy fluxes at the atmosphere-surface interface, provides insights into the magnitude of the Arctic amplification irrespective of the sources of the amplification. This is possible by comparing the sensitivities of the surface temperature change, described by changes in the upward longwave irradiance, between the Arctic and the global average. In this way, one can relate the magnitude of the amplification to the sensitivity of surface temperature to changes in the various energy flux terms in the surface energy balance equation.
In winter, the outgoing longwave radiation at the top of the atmosphere over the polar cap north of 708N is balanced primarily by the sum of meridional heat transport and the surface to air heat flux. The conductive heating is estimated to be about 20%-30% of the magnitude of the meridional transport heating (Serreze and Barry 2005). The temperature in the free troposphere is a result of a balance between longwave radiative cooling and subsidence and horizontal convergence heating. The mean subsidence is connected by the mass continuity constraint to a net northward potential energy flux into the polar cap troposphere (Serreze and Barry 2005). Average temperature soundings in the High Arctic winter reveal a strong temperature inversion that extends from the surface to the over 1 km in height (e.g., Zhang et al. 2011). This is commonly referred to as a surface-based inversion (e.g., Serreze et al. 1993; Hudson and Brandt 2005; Bourne et al. 2010; Zhang et al. 2011), although the term is best applied in the climatological mean since on short time scales mixing due to strong surface winds or open water in leads and polynya can result in a nearly adiabatic surface layer, which lifts the base of the inversion above the surface. Since the subsidence heating and warm air horizontal advection must go to zero at the surface, the longwave cooling of the surface and the overlaying air allows the surface air temperatures to cool below values found in the lower free troposphere. The result is the formation of a very stable boundary layer, which contains, in the climatological sense, a surface-based temperature inversion.
Of particular interest in the current work is the role of a stable boundary layer in reducing the upward vertical sensible and latent heat fluxes, thereby enhancing the amplification of surface warming. The inversion is not the driver of the amplification but rather a factor in determining the sensitivity of the surface temperature response. The drivers are given in the aforementioned list of the causes of amplification, which include changes in sea ice (e.g., Screen and Simmonds 2010b), large-scale dynamics (e.g., Yang et al. 2010), and clouds (Shupe and Intrieri 2004).
The boundary layer static stability has a profound effect on the surface vertical fluxes that determine how the surface temperature responds to perturbations in the forcings (Busch et al. 1982; Persson et al. 2002; Serreze and Francis 2006). Medeiros et al. (2011) showed how a bias in the Arctic inversion strength as simulated in climate models impacted the surface energy budget with consequences on the surface temperature response. Pavelsky et al. (2011) showed how the different surface heat fluxes over sea ice compared to land will impact their respective inversion strengths. Bintanja et al. (2012) used a global climate model to show that artificially reducing the boundary layer mixing resulted in a stronger Arctic amplification signal. Their results suggest that some of the intermodel differences in predicted amplification strength are related to model differences in surface-based inversion strengths.
The Arctic boundary layer often consists of a surfacebased temperature inversion that is most frequent and most stable during the winter (Curry 1983; Kahl 1990; Bradley and Keimig 1992; Serreze and Barry 2005; Lesins et al. 2010; Devasthale et al. 2010; Bourne et al. 2010; Zhang et al. 2011). The lack of solar irradiance in the winter coupled with net longwave cooling of the surface combine to create inversions that commonly exceed 1 km in depth and 208C in magnitude at some High Arctic stations (Lesins et al. 2010).Acritical aspect relevant to the amplification issue is that the inversion layer dynamically decouples the surface from the free troposphere (Garratt 1992; Hartmann 1994). The inhibition of turbulent mixing prevents the surface heat flux perturbations from easily spreading throughout the troposphere, thereby concentrating the response close to the surface and resulting in an amplification of the warming. The surface energy balance equation can be used to link an energy perturbation at the surface with the magnitude of the temperature response. Since the surface energy balance is a statement of energy conservation that is applicable on all time scales, this framework can be used to address the full range of fluctuations from daily to climate scales. Lu and Cai (2009) used a perturbation surface heat budget equation to interpret changes in the surface albedo feedback and cloud radiative forcing derived from a global climate model.
A description of the datasets and sites used is given in section 2. This is followed in section 3 by a presentation of the observed warming trends at the Arctic stations and evidence that the surface-based temperature inversion and the surface temperature play important roles in the amplification. In section 4, a surface energy balance approach is provided to illustrate how changes in the various vertical energy fluxes impact the temperature amplification. Section 5 contains a discussion of the observations using the surface energy balance framework. A summary follows in section 6.
2. Dataset and methods
This study uses the upper-air radiosonde record for all Canadian stations that start no later than 1970 from the Integrated Global Radiosonde Archive (IGRA), which contains individual soundings from the global radiosonde network (Durre et al. 2006). The archive underwent automated quality assurance procedures for the temperature soundings (Durre et al. 2008). We use the temperatures from the ''Derived2'' version of the archive in which the formatting has been modified for ease of use (Durre and Yin 2008).
The time series of surface temperatures and the temperature at the first elevated level above the ground are used to determine the vertical temperature difference from which the surface-based temperature lapse rate is computed. The first value in the IGRA sounding is the surface air temperature. This is done for 22 upperair stations in the Canadian Arctic whose locations are shown in Fig. 1. These stations have similar record lengths, used the same rawinsonde instrument packages, and were subjected to same quality control, which helps to reduce difficulties that can arise from comparing different sites. The vast geographic area covered by the Canadian Arctic provides a wide range of meteorological conditions including stations that are located on or away from the Arctic coast where sea ice has an influence.
The 22 upper-air stations met the following criteria: 1) the record had to start before the beginning of 1971, 2) there could not be any time gaps in the data of more than 2 months, and 3) the record had to extend to the end of 2010 providing a 40-yr time series. Station names, locations, elevations, starting year, and the year when a consistent warming trend started based on the 5-yr running means of the surface air temperature are listed in Table 1.
3. Surface warming trends
Figure 2 shows the time series of the annual-mean surface air temperature averaged over all 22 stations superimposed on the annual means for the individual stations. A warming trend is evident in spite of the large interannual variability. The warming trend averaged over all the CanadianArctic stations from 1971 to 2010 is 0.698 6 0.138Cdecade21 and from1991 to 2010 is 1.068 6 0.358C, which corresponds to amplification factors of about 4.1 and 5.0 when compared against the global averaged (land and ocean) trend of 0.178 and 0.218C decade21 during the same time periods derived from GISS data (Hansen et al. 2006).
The amplification factors for the annual-mean temperature trends from 1971 to 2010 and from 1991 to 2010 for each Canadian station are shown in Fig. 3 as a function of the station latitude. The amplification factor is computed as the ratio of the station's annual-mean warming trend to the GISS global average warming trend for the same time period. The linear regression of the 40-yr trends shows an increase in the amplification factor with latitude from about 3.1 at 508Nto about 5.5 at 808N. The latitudinal increase in amplification over the latest 20 yr is accompanied by more scatter. Much of the scatter is a result of regional variations in the warming trend, in particular differences between the western and eastern Canadian Arctic.
Figure 3 shows that Arctic amplification extends into the midlatitudes, at least to 478N, in areas quite distant from the Arctic and Arctic sea ice. The figure uses an open symbol for stations that are located close to sea ice that occurs during some part of the year, while the closed symbol denotes stations far away from any sea ice. The average amplification factors for the 1971 to 2010 period for stations close to and far from sea ice are 4.960.5 and 3.2 6 1.2, respectively. These increase to 6.5 6 2.4 and 3.6 6 2.4, respectively, for the 20-yr period from 1991 to 2010. The values are consistent with the interpretation that melting sea ice plays a role in the magnitude of the Arctic amplification (e.g., Chapin et al. 2005).
The warming trend from 1971 to 2010 and from 1991 to 2010 for each station is given by month in Fig. 4. The reduced scatter for the 40-yr trends shows more clearly that the warm season months from May to August have smaller warming trends than the cold season months. In spite of the larger scatter in the 20-yr trends, the basic conclusion remains the same that the winter and autumn months exhibit the greatest amplification. These observations are consistent with the monthly temperature trends from 1989 to 2009 by Screen and Simmonds (2010a) for all meteorological stations north of 708N. The warming trends reported here for the surface air temperature are similar to the Advanced Very High Resolution Radiometer-(AVHRR) derived skin temperature trends from 1981 to 2003 by Comiso (2003) and reanalyzed by Serreze and Francis (2006).
The monthly warming trends averaged over all stations are plotted as a function of surface-based inversion strength for 40- and 20-yr periods in Figs. 5a,b, respectively. The inversion strength is computed from individual soundings as the temperature difference between the first elevated temperature measurement and the surface, normalized to a 1-km-thick layer, and then averaged on a monthly basis over the given time period. A positive (negative) value for the surface-based inversion strength means the air temperature is increasing (decreasing) with height starting with the surface temperature, which by normal convention in meteorology is described as a negative (positive) lapse rate. Since the first elevated sounding layer has a warming trend very similar to the surface trend, no significant trends in the inversion strengths were detected.
A concern with the computed inversion strength is whether the calculation up to the first elevated sounding level is representative of the inversion strength right at the surface where the surface sensible and latent heat fluxes are determined. For the IGRA soundings, the first sounding altitude varies from 100 to 200 m in the winter and from 150 to 400 m in the summer. The IGRA archive reports significant and mandatory levels, and hence the first height will vary within these ranges depending on the detailed behavior of the vertical temperature structure.
Surface-based inversions created by longwave cooling from the ground are strongest at the ground and become weaker with height (e.g., Mahesh et al. 1997). Based on a detailed analysis of Environment Canada soundings at Eureka from 1984 to 2007 (not shown), the difference between the 100- and 200-m-thick layer lapse rate is less than 10% in the winter but becomes 50% when comparing 100 to 1000 m. Therefore, the variation in the first sounding level in IGRA does not create a significant bias. However, our reported inversion strengths are probably a bit weaker than the actual inversion strength right at the surface. This would simply shiftthe points in Figs. 5a,b and 6a,b to the right without changing either the character of the plots or our conclusions.
The time lag associated with the temperature response in a rapidly rising radiosonde results in a measured lapse rate smaller than the actual value (Mahesh et al. 1997). Mahesh et al. (1997) applied this correction to the monthly-mean sounding at Fairbanks (658N) for January 1989 and found that in the lowest 100 m an additional stability of 108C km21 had to be added to the measured stability of about 308C km21. This bias is in addition to the underestimation of the surface lapse rate due to the elevated measurement of the first sounding height discussed above but would also result in a shiftof the points to the right in Figs. 5a,b and 6a,b and so would not alter our conclusions.
The warming trends are generally greatest during the winter half of the year when the surface inversion strengths are larger. The 20-yr trend for March stands out as outside of this pattern. Closer examination (not shown) reveals that from 1991 to 2010 the average March temperatures underwent a strong cooling trend for stations in northwest Canada. The monthly warming trends averaged over all stations are plotted as a function of the monthlymean surface temperatures for 40- and 20-yr periods in Figs. 5c,d, respectively. The pattern is very similar to that seen with the inversion strength (Figs. 5a,b), with the cold months experiencing the larger amplification. Furthermore, since inversion strength generally increases with latitude while the temperature decreases with latitude, the latitudinal dependence of amplification as seen in Fig. 3 is reinforced.
The Fig. 6 panels are the same as Fig. 5, except that the winter and summer averages are plotted for each station with straight lines joining them. For the purposes of this figure, the summer months are defined as May-October with the other six months included in the winter average. This division is based on the monthly averaged magnitudes of the inversion strengths and surface temperatures. The clear separation of the warming trend by season is linked to the roles of the surface-based inversion and the average surface air temperature as predicted by the SEB equation, which will be elaborated upon in the following section.
4. Surface energy balance approach
The SEB equation states that the sum of all the vertical energy fluxes evaluated at the air-surface interface must equal zero,
where S and I are the solar and longwave irradiances, Fsh and Flh are the sensible and latent heat fluxes at the surface, C is the heat conduction from below the surface, and the subscripts d and u denote downward and upward directions. All energy fluxes are defined to be positive if they are directed toward the surface. Note that this form of energy balance does not include a time derivative storage term since we are assuming an interface of zero heat capacity that is infinitesimally thin.
Writing the upward solar irradiance as the product of the downward solar irradiance and the directionally averaged surface albedo a and assuming the surface has an average longwave broadband emissivity «, Eq. (1) becomes
where Ts is the skin temperature and s is the Stefan- Boltzmann constant. After taking the differential of Eq. (2) and solving for the change in the skin temperature DTs, we obtain
is defined as the change in the downward net irradiance at the surface, where it is assumed for convenience that changes in the upward solar irradiance due to surface albedo changes can be absorbed into the definition. Equation (3) states that the temperature responds to changes in the downward longwave irradiance DId, downward solar irradiance DSd, surface albedo Da, sensible heat flux DFsh, latent DFlh heat flux, and heat conduction from below the surface DC.
Since there is a rapid increase in wind speed with height in the surface boundary layer, the magnitude of the horizontal temperature advection in the surface boundary layer will be rapidly reduced in approaching the ground. At the surface interface, the horizontal advection of sensible heat is zero; however,when the atmospheric column above the surface experiences a temperature change due to horizontal sensible heat advection its effects are transmitted to the surface by changes in the vertical sensible turbulent flux and in the downward longwave irradiance. By this indirect route, the effects of meridional heat transport, typically evaluated over extended layers bounded by the mandatory pressure levels, are included in the calculation of temperature change using SEB.
The surface energy balance equation requires the skin temperature instead of the measured surface air temperature, which is typically at the screen height about 1.5 m above the surface. The height difference between the ground and the screen is often accompanied by a temperature difference. The difference is smallest when turbulent mixing is strong and when skies are overcast so that the surface does not experience intense daytime solar heating or nocturnal longwave cooling. The skin temperature can be related to the surface air temperature by using an appropriate heat transfer model for the surface layer or by imposing an empirically determined temperature difference (Overland and Guest 1991). We will assume, based on preliminary analysis discussed in section 5, that there are no trends in the skin to air temperature difference over the time periods considered here. Future research is needed to test whether trends in cloud cover or boundary layer turbulence are significant enough to induce a trend in the skin to air temperature difference.
Based on the SEB Eq. (3), Arctic amplification has potentially four contributing factors: 1) a larger change in net downward radiation at the Arctic surface compared to the global average, 2) a larger conductive heat flux change than the global average, 3) different sensible and latent heat flux responses that result in a larger surface temperature response, and 4) a colder skin temperature compared to the global average as a result of the temperature term in the denominator of Eq. (3). We now consider examples how each of these factors might influence the amplification factor.
In the first factor the Arctic surface can be experiencing a larger net downward radiation change than the global average. For example, an increase in the meridional energy and water vapor transport to the Arctic, as a result of climate change, can warm and moisten the troposphere above the Arctic resulting in an increase in downward longwave compared to the global average. Also, the retreat of the sea ice margin will promote warming of the overlaying air, which can advect farther north and enhance the downward longwave irradiance (Serreze et al. 2009; Higgins and Cassano 2009).
In the second factor, the conductive flux through the sea ice and snow to the surface will increase over the Arctic Ocean as the sea ice becomes thinner, a feedback caused by the regional warming. Since this feedback is not occurring outside of the polar region it can contribute to part of the amplification in the Arctic.
For the third factor, consider a hypothetical situation where the global average and the Arctic both experience the same increase in net downward radiation at the surface. The temperature response will differ because of the role static stability plays in the boundary layer. Since the globally averaged surface layer is typically well mixed during the daytime, the response to the increase in net downward radiation at the surface will be an increase in the upward sensible and latent heat fluxes, which will spread the warming over a deeper layer resulting in a reduced surface air temperature response. In the winter Arctic, the sensible and latent heat fluxes will be nearly zero because of the high static stability. Consequently, all the response to the increased downward radiation must be compensated by the upward longwave irradiance resulting in a larger amount of surface warming. The amplification is due to the lack of a sensible and latent heat flux response.
In situations where the surface layer wind shear is strong enough to overcome the static stability and induce a sensible heat flux, the heat flux will be directed toward the surface because the boundary layer temperature is increasing with height. In this case, changes in the boundary layer wind speed, which is transmitted to the surface by the wind shear, will alter the sensible heat flux and thereby invoke a surface temperature response, which will impact the amplification.
The higher sensitivity of the surface temperature response in the presence of an inversion can be seen at nonpolar latitudes when the sun rises after a clear, calm night during which a nocturnal stable boundary layer developed (Garratt 1992; Hartmann 1994). The initial solar heating of the surface is transmitted by conduction and very weak convection to only a thin layer of air closest to the ground because of the high static stability that inhibits turbulent mixing. As a result, the surface air temperature increases rapidly since the solar heating is confined to a thin atmospheric layer with relatively low heat capacity. As the temperature inversion erodes from below and a deeper well-mixed boundary layer is established, the air temperature increases more slowly since the solar heating is now distributed through a deeper well-mixed layer, which continues to grow in depth during the course of the morning. The influence of a surface-based inversion also helps to explain why the globally average diurnal minimum temperature is rising more rapidly than the diurnal maximum temperature (Vose et al. 2005).
For the fourth factor consider the situation where the increase in surface net downward radiation and the sensible and latent heat feedbacks are globally uniform. Even in this situation, the Arctic will experience amplification because of the power law dependence of the compensating upward longwave irradiance on the skin temperature. For colder skin temperatures, the amount of surface warming must be larger in order to reestablish surface energy balance.
The SEB approach in characterizing Arctic amplification as described in section 4 revealed four contributing factors. Two of these factors, boundary layer stability in suppressing the upward flux of sensible and latent heat and colder skin temperatures in eliciting a larger warming to balance any increase in net downward radiation, are consistent with the observations presented in Figs. 5 and 6. The other two amplification factors, changes in the surface net downward radiation and changes in the subsurface conductive flux, could not be determined using the radiosondes and will require future work.
SEB applies right at the air-surface interface using the skin temperature, whereas the temperature measurements are taken about 1.5 mabove the ground in the air. The skin-air temperature difference can be significant when the surface experiences large net radiative heating or cooling. The MODIS skin temperature product was examined for Eureka and Resolute for the time period 2004-09 and compared to the surface air temperature measurements from IGRA (Pike-Thackray 2011). The MODIS skin temperatures are retrieved for clear skies when the net upward longwave radiation at the surface is the largest, giving the greatest difference between the skin and air temperatures. Under clear skies, in the absence of sunlight, the average winter monthly skin temperature was 38-68C colder than the surface air temperature and, more importantly, there were no discernible trends in the skin-air temperature difference from 2004 to 2009 (Pike-Thackray 2011). Hence, we have some justification in using the surface air temperature trend as a proxy for the skin temperature trend in the SEB equation.
Figure 5 shows that both the inversion strength and surface air temperature give very similar dependences on the warming trend when plotted as monthly averages over all 22 stations. This is not surprising because the inversion strength is in large part determined by how cold the surface temperature is (Bourne et al. 2010). The coldest temperatures and strongest inversions both occur in the winter months. As a result of this correlation, it is difficult to quantitatively separate out the effects of the inversion from those of the temperature as predicted in the SEB approach.
Figure 6 simplifies the analysis by partitioning the year into two seasons instead of 12 months as shown in Fig. 5. The winter versus summer difference in the dependence of the warming trend is shown for each station individually. For the time period 1971-2010, the winter to summer differences exhibit a more consistent and linear pattern for the surface air temperature influence shown in Fig. 6c compared to the inversion strength influence shown in Fig. 6a. The same statement cannot be made for the period 1991-2010 shown in Figs. 6b,d because of the larger scatter of points, which is related to the problems introduced by the shorter time period. The more consistent pattern in Fig. 6c suggests that the surface temperature factor of amplification is important in explaining much of the interstation and interseasonal differences in the warming trend.
It is possible to roughly estimate the magnitude of the amplifying effect from the skin temperature dependence on the upward longwave response since radiation varies with the cube of the ratio of the warm to cold surface temperatures. Choosing typical surface temperature values to represent the globe (288 K) and the High Arctic winter (240 K) gives a global to Arctic ratio of the denominators from Eq. (3) of (288/240)3 ' 1.7, which explains a portion of the Arctic amplification.
The surface air temperature effect on amplification is given by the upward longwave response, which varies in a continuous fashion following the cubic relationship with temperature as shown in Eq. (3). The same continuous behavior is not expected for the amplification dependence on sensible and latent heat fluxes via the surface layer static stability. The influence of static stability in suppressing the upward sensible and latent heat fluxes will operate as the static stability is varied from unstable to neutral conditions. Once the boundary layer becomes weakly stable, the direction of the sensible heat fluxwill reverse and be directed toward the surface, which will aid in the amplification. As the stability continues to increase or the wind shear decreases, the Richardson number criterion for mechanical turbulence will no longer be met, the turbulence and resulting sensible heat flux will be suppressed, and the amplification will no longer be sensitive to increases in static stability. Figure 6a shows some support for this explanation as most of the increase in the warming trend takes place for inversion strengths less than 108C km21.
The SEB approach along with the measurements presented here support the following explanation: as one moves farther into the Arctic the static stability in the boundary layer will eventually be strong enough to suppress the sensible and latent heat response to changes in the surface net downward radiation, resulting in a greater sensitivity of the surface temperature response. As one continues to move deeper into the High Arctic, the boundary layer inversion plays a less important role and instead the colder surface temperatures help to increase the amplification factor. The reversal of the sensible heat flux direction may also contribute to amplification. This explanation does not account for any latitudinal dependence in the surface net downward radiation change. For example, amplification associated with sea ice loss via the net downward radiation and conduction factors given in Eq. (3) are not examined in this study.
Figure 7 is a schematic diagram summarizing the qualitative changes in vertical energy fluxes with global warming. The response to an increase in net downward radiation at the surface, required to maintain SEB, is an increase in the surface air temperature, closely linked to the skin temperature, which increases the upward longwave irradiance. However, since the globally averaged daytime boundary layer is typically near neutral stability, the response also includes an increase in the upward latent and sensible heat fluxes. The increase in latent heat flux drives a stronger hydrological cycle and is also responsible for the strong convective heating response in the upper tropical troposphere, atmospheric consequences of the increasing atmospheric carbon dioxide (Solomon et al. 2007).
In the Arctic winter, on the other hand, the stable boundary layer confines the heating from the positive radiative forcing to remain close to the surface resulting in an amplified surface temperature increase. Polynyas and open leads in the sea ice during the winter will create large fluxes of sensible and latent heat, which can complicate the amplification response that we observe over the land stations analyzed in this work. During the Arctic summer, much of the additional net downward radiation at the surface is used to melt sea ice and snow and possibly to increase the upward sensible and latent heat fluxes to the atmosphere due to the weaker static stability at this time of year. Furthermore, the colder surface temperature elicits a larger temperature change in order to balance the increase in net downward radiation. This effect helps to explain the increase in amplification with latitude into the High Arctic.
From 22 Canadian radiosonde stations with nearly complete records extending from 1971 to 2010, the average surface air temperature warming trend was measured to be 0.698 6 0.138C decade21. The Arctic amplification factors for individual stations ranged from 1.4 to 5.2 when the warming trends are compared to the global average of 0.178C decade21. The amplification factor exhibits a strong latitudinal dependence varying from 2.6 to 5.2 as the latitude increases from 508 to 808N. The results show that significant amplification is also occurring in the midlatitudes in Canada and that the amplification is enhanced for stations located close to sea ice. The warming trend exhibits a strong seasonal dependence with the least warming taking place during the summer months. The monthly variations in the warming trend were shown to be closely related to both the surface-based inversion strength and the mean monthly surface temperatures.
Based on the surface energy balance equation, there are four contributing factors available to explain Arctic amplification: 1) a larger change in net downward radiation at theArctic surface compared to the global average, 2) a larger conductive heat flux change than the global average, 3) weaker sensible and latent heat flux responses that result in a larger surface temperature response in the Arctic, and 4) a colder skin temperature compared to the global average. This study was able to examine the latter two using the radiosonde measurements.
By associating a weaker sensible and latent heat flux response to the higher static stability in the Arctic surface layer, we can conclude that the observed relationship between the warming trends and the inversion strength is consistent with the SEB analysis. However, the effect of a colder skin temperature in eliciting a larger temperature response to changes in the net downward radiation at the surface is also consistent with the SEB analysis. Both of these factors play a role in explaining the magnitude of the amplification factor.
In summary, the measurements and SEB approach suggest that the magnitude of the amplification is determined in part by two factors: 1) the static stability of the boundary layer increases sufficiently to switch offthe response of the latent and sensible heat fluxes and 2) the colder surface temperatures with latitude into the High Arctic elicit a stronger temperature response because of its dependence on the upward longwave irradiance. Quantifying the relative contributions from all four factors of amplification and determining the magnitudes of the various proposed forcing mechanisms require models to assess the forcing changes and responses. These are topics for continuing investigations.
Acknowledgments. Financial support for this research was provided by the Canadian Foundation for Climate and Atmospheric Research (CFCAS), the Canadian Foundation for Innovation (CFI), the National Science and Engineering Research Council (NSERC) of Canada, the Canadian Space Agency (CSA), the Ontario Innovation Trust (OIT), theOntarioResearch Fund (ORF), theNova ScotiaResearch and Innovation Trust (NSRIT), and the Government of Canada International Polar Year fund. We also thank Environment Canada, in particular KazHiguchi and Shannon Allen, formaking available the archived surface and radiosonde measurements. Thanks also to Colin Pike-Thackray for help in computing the skin-air temperature differences. We thank the National Climatic Data Center of NOAA for making the IGRA dataset publicly available. Finally we appreciate the useful comments from the anonymous reviewers.
Alexeev, V. A., P. L. Langen, and J. R. Bates, 2005: Polar amplification of surface warming on an aquaplanet in ''ghost forcing'' experiments without sea ice feedbacks. Climate Dyn., 24, 655- 666, doi:10.1007/s00382-005-0018-3.
Bintanja, R., E. C. van der Linden, and W. Hazeleger, 2012: Boundary layer stability and Arctic climate change: A feedback study using EC-Earth. Climate Dyn., 39, 2659-2673, doi:10.1007/s00382-011-1272-1.
Bourne, S.M.,U. S. Bhatt, J. Zhang, and R. Thoman, 2010: Surfacebased temperature inversions in Alaska from a climate perspective. Atmos. Res., 95, 353-366.
Bradley, R. S., and F. T. Keimig, 1992: Climatology of surfacebased inversions in the North American Arctic. J. Geophys. Res., 97 (D14), 15 699-15 712.
Busch, N., U. Ebel, H. Kraus, and E. Schaller, 1982: The structure of the subpolar inversion-capped ABL. Meteor. Atmos. Phys., 31, 1-18.
Chapin, F. S., III, and Coauthors, 2005: Role of land-surface changes in Arctic summer warming. Science, 310, 657-660.
Comiso, J. C., 2003: Warming trends in the Arctic from clear-sky satellite observations. J. Climate, 16, 3498-3510.
_____, C. L. Parkinson, R. Gersten, and L. Stock, 2008: Accelerated decline in the Arctic sea ice cover. Geophys. Res. Lett., 35, L01703, doi:10.1029/2007GL031972.
Curry, J., 1983: On the formation of continental polar air. J. Atmos. Sci., 40, 2278-2292.
Devasthale, A., U. Willen, K.-G. Karlsson, and C. G. Jones, 2010: Quantifying the clear-sky temperature inversion frequency and strength over the Arctic Ocean during summer and winter seasons fromAIRS profiles. Atmos.Chem. Phys., 10, 5565-5572.
Durre, I., and X. Yin, 2008: Enhanced radiosonde data for studies of vertical structure. Bull. Amer. Meteor. Soc., 89, 1257-1262.
_____, R. S. Vose, and D. B. Wuertz, 2006: Overview of the Integrated Global Radiosonde Archive. J. Climate, 19, 53-68.
_____,_____, and _____, 2008: Robust automated quality assurance of radiosonde temperatures. J. Appl. Meteor. Climatol., 47, 2081- 2095.
Garratt, J. R., 1992: The Atmospheric Boundary Layer. Cambridge University Press, 316 pp.
Garrett, T. J., and C. Zhao, 2006: Increased Arctic cloud longwave emissivity associated with pollution from mid-latitudes. Nature, 440, 787-789.
_____, L. F. Radke, and P. V. Hobbs, 2002: Aerosol effects on cloud emissivity and surface longwave heating in the Arctic. J. Atmos. Sci., 59, 769-778.
Graversen, R. G., and M. Wang, 2009: Polar amplification in a coupled climate model with locked albedo. Climate Dyn., 33, 629-643, doi:10.1007/s00382-009-0535-6.
Hansen, J., M. Sato, R. Ruedy, K. Lo, D. W. Lea, and M. Medina- Elizade, 2006: Global temperature change. Proc. Natl. Acad. Sci. USA, 103, 14 288-14 293.
Hartmann, D. L., 1994: Global Physical Climatology. Academic Press, 411 pp.
Higgins, M. E., and J. J. Cassano, 2009: Impacts of reduced sea ice on winter Arctic atmospheric circulation, precipitation and temperature. J. Geophys. Res., 114, D16107, doi:10.1029/ 2009JD011884.
Holland, M. M., and C. M. Bitz, 2003: Polar amplification of climate change in coupled models. Climate Dyn., 21, 221-232.
Hudson, S. R., and R. E. Brandt, 2005: A look at the surface-based temperature inversion on the Antarctic Plateau. J. Climate, 18, 1673-1696.
Kahl, J. D., 1990: Characteristics of the low-level temperature inversion along the Alaskan Arctic coast. Int. J. Climatol., 10, 537-548.
Lesins,G., T. J.Duck, and J. R. Drummond, 2010: Climate trends at Eureka in the CanadianHighArctic. Atmos.-Ocean, 48, 59-80.
Liu, Y., J. R. Key, and X. Wang, 2008: The influence of changes in cloud cover on recent surface temperature trends in the Arctic. J. Climate, 21, 705-715.
Lu, J., and M. Cai, 2009: Seasonality of polar surface warming amplification in climate simulations. Geophys. Res. Lett., 36, L16704, doi:10.1029/2009GL040133.
_____, and _____, 2010: Quantifying contributions to polar warming amplification in an idealized coupled general circulation model. Climate Dyn., 34, 669-687.
Mahesh, A., V. P. Walden, and S. G. Warren, 1997: Radiosonde temperature measurements in strong inversions: Correction for thermal lag based on an experiment at the South Pole. J. Atmos. Oceanic Technol., 14, 45-53.
Manabe, S., and R. J. Stouffer, 1980: Sensitivity of a global climate model to an increase of CO2 concentration in the atmosphere. J. Geophys. Res., 85 (C10), 5529-5554.
Mauritsen, T., and Coauthors, 2011: An Arctic CCN-limited cloudaerosol regime. Atmos. Chem. Phys., 11, 165-173.
Medeiros, B., C. Deser, R. A. Tomas, and J. E. Kay, 2011: Arctic inversion strength in climate models. J. Climate, 24, 4733-4740.
Overland, J. E., and P. S. Guest, 1991: The snow and air temperature budget over sea ice during winter. J. Geophys. Res., 96 (C3), 4651-4662.
Pavelsky, T. M., J. Boe, A. Hall, and E. J. Fetzer, 2011: Atmospheric inversion strength over polar oceans in winter regulated by sea ice. Climate Dyn., 36, 945-955.
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, doi:10.1029/2000JC000705.
Pike-Thackray, C., 2011: Reconstruction of High Arctic winter surface energy fluxes. M.S. thesis, Dept. of Physics and Atmospheric Science, Dalhousie University, 95 pp.
Polyakov, I. V., and Coauthors, 2002: Observationally based assessment of polar amplification of global warming. Geophys. Res. Lett., 29, 1878, doi:10.1029/2001GL011111.
Screen, J. A., and I. Simmonds, 2010a: Increasing fall-winter energy loss from the Arctic Ocean and its role in Arctic temperature amplification. Geophys. Res. Lett., 37, L16707, doi:10.1029/ 2010GL044136.
_____, and _____, 2010b: The central role of diminishing sea ice in recent Arctic temperature amplification. Nature, 464, 1334-1337.
Serreze, M. C., and R. G. Barry, 2005: The Arctic Climate System. Cambridge University Press, 385 pp.
_____, and J. A. Francis, 2006: The Arctic amplification debate. Climatic Change, 76, 241-264.
_____, and R. G. Barry, 2011: Processes and impacts of Arctic amplification: A research synthesis. Global Planet. Change, 77, 85-96.
_____, J. E. Box, R. G. Barry, and J. E. Walsh, 1993: Characteristics of Arctic synoptic activity, 1952-1989. Meteor. Atmos. Phys., 51, 147-164.
_____,A. P. Barrett, J. C. Stroeve, D. N. Kindig, and M. M. Holland, 2009: The emergence of surface-based Arctic amplification. Cryosphere, 3, 11-19.
Shindell, D., and G. Faluvegi, 2009: Climate response to regional radiative forcing during the twentieth century. Nat. Geosci., 2, 294-300.
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, 616-628.
Solomon, S., D. Qin, M. Manning,M. Marquis, K. Averyt,M.M. B. Tignor, H. L. Miller Jr., and Z. Chen, Eds., 2007: Climate Change 2007: The Physical Science Basis. Cambridge University Press, 996 pp.
Vose, R. S., D. R. Easterling, and B. Gleason, 2005: Maximum and minimum temperature trends for the globe: An update through 2004. Geophys. Res. Lett., 32, L23822, doi:10.1029/ 2005GL024379.
Yang, X.-Y., J. C. Fyfe, and G. M. Flato, 2010: The role of poleward energy transport in Arctic temperature evolution. Geophys. Res. Lett., 37, L14803, doi:10.1029/2010GL043934.
Zhang, Y., D. J. Seidel, J.-C. Golaz, C. Deser, and R. A. Tomas, 2011: Climatological characteristics of Arctic and Antarctic surface-based inversions. J. Climate, 24, 5167-5186.
GLEN LESINS, THOMAS J. DUCK, AND JAMES R. DRUMMOND
Department of Physics and Atmospheric Science, Dalhousie University, Halifax, Nova Scotia, Canada
(Manuscript received 5 December 2011, in final form 7 May 2012)
Corresponding author address: Glen Lesins, Department of Physics and Atmospheric Science, Dalhousie University, Halifax NS B3H 3J5, Canada.
(c) 2012 American Meteorological Society
[ Back To Technology News's Homepage ]