ETH Greenland Summit Experiment

This is not an official web page of the project, but a personal overview over some of the work that has been done in the project.

Introduction

The Greenland Summit Environmental Observatory (GEO-Summit, Summit Camp) lies in the center of the Dry Snow Zone of the Greenland Ice Sheet at  72°35' N, 34°30' W at an elevation of 3230 m a.s.l.
 
Due to the importance of the Dry Snow Zone for the mass balance of the entire Greenland Ice Sheet,  detailed information on the energy balance is needed to understand the role of Greenland in a changing global climate.

Besides the energy balance study at this representative site, fundamental research was conducted on longwave radiation exchange within the boundary layer and on the spectral reflectivity of snow.

Energy Balance Studies

All components of the surface energy balance were investigated during an intensive project season which lasted from May 2001 to July 2002.

Net Radiation is determined from the individual measurements of the components of the radiation balance measured on the radiation monitoring tower (Figure 1).
radiation monitoring tower at Summit Greenland (2002) Figure 1: Radiation monitoring at Summit, Greenland.

The annual cycle of the radiation balance at Summit, Greenland, for 2001  is shown in Figure 2.
radiation balance summit, Greenland, 2001 Figure 2: Annual cycle of radiative quantities observed at Summit, Greenland.

Temperatures measured on a chain of thermistors and thermocouples allow the calculation of the Subsurface Heat Flux. The deepest measurement level was installed at a depth of 15m. Monthly mean tautochrones are shown in Figure 3.
tautochrones, Summit Figure 3: Tautochrones observed at Summit, Greenland.
  


Sensible Heat Flux  was directly measured with a eddy correlation system at 4 levels, but wind speeds and air temperatures measured at 8 levels of the 50 m meteorological tower (Figure 4) are available for calculations using the gradient method.


50 m meteorological tower at Summit, Greenland Figure 4: The 50 m meteorological tower at Summit, Greenland.

Similarly, Latent Heat Flux is directly measured or determined from measurements at different levels. With the use of snow lysimeters sublimation and  resublimation is monitored regularly. A comparison of the latent heat flux determined from the lysimeter observation with calculations using the profile method is shown in Figure 5.
latent heat flux: lysimeter vs profile method Figure 5: Latent Heat Flux observed with lysimeter and calculated using the profile method, Summit, Greenland.

As an example, the monthly mean diurnal cycles of the different components of the energy balance are shown in Figure 6.
energy balance, Summit, Greenland, Feb2002 energy balance, Summit, Greenland, Jun2002
Figure 6: Diurnal cycles of the components of the energy balance at Summit, Greenland, for  February 2002 (left) and June 2002 (right).

From the observations during the ETH Summit Project we can conclude:
  • During the winter months, the surface is cooled due to a negative longwave radiation balance. This cooling effect is mainly balanced by a positive sensible heat flux, and to approximately 1/3 by the subsurface heat flux. Latent heat flux is slightly positive. As the sun is below the horizon between mid November and late January, no significant diurnal variation of energy fluxes occur.
  • In summer, monthly mean net radiation is positive, and net radiation is positive throughout the larger part of the day. Monthly mean sensible heat flux is close to zero, but a transport of sensible heat away from the surface can be seen around solar noon as early as in April. Negative sensible flux dominates for 10 hours in June, indicating unstable conditions close to the surface.
  • Positive net radiation is balanced in equal parts by the cooling due to sublimation (loss of latent heat) and by a negative subsurface heat flux. Warming of the surface by a positive latent heat flux (resublimation) can be observed during clear-sky summer nights, but in the monthly mean diurnal cycle, latent heat flux remains directed away from the surface.
  • The residuals of the seasonal energy balance are relatively large, ranging from -6 Wm-2 in fall to 4 Wm-2 in spring.

Radiative Flux Divergence

The frequent occurrence of strong temperature inversions makes Summit a field laboratory for studying the stable boundary layer (SBL). Although recognized as an important component of the thermodynamics of the boundary layer (Robinson, 1950; Garratt and Brost, 1981), the divergence of the longwave radiative flux has never been investigated experimentally across several layers within the lower boundary layer. The few available observations are either limited to a single layer near the ground, or cover only a limited period of time. Especially the shape of the radiative heating/cooling rate profile has been under debate (Stull, 1988). It has so far only been determined from radiative transfer calculations. The results of such calculations, however, have been contradictory for the near surface layers, as they are highly sensitive to vertical resolution (Räisänen, 1996).

Observations

During the ETH Greenland Summit Experiment (June 2001 to July 2002)  the profile of longwave fluxes was measured on up to 6 levels of the 50 m meteorological tower (Figure 4). A relative calibration of the pyrgeometers (Eppley Precision Infrared Radiometer) reduced the uncertainty of net flux difference measurements to ± 0.75 Wm-2 and a correction for tower‘s influence on the longwave flux measurements was applied (Hoch et al., 2007). As a first example, the longwave fluxes measured at 2 m and 50 m are shown for 21 January 2002 (Figure 7, left panel). The divergence of the net longwave flux of (0.35 Wm-3) corresponds to a longwave radiative cooling of about -30 K per day (K d-1) . The divergence of the longwave radiative fluxes is shown in the right panel of Figure 7.
lw fluxes 21 jan 2002, Summit, Greenland lw divergence 21 Jan 2002, Summit, Greenland
Figure 7: Longwave radiative fluxes measured on 21 January , 2002, at Summit, Greenland, at 2 m and 50 m above the surface (left panel), and the corresponding  radiative flux divergence (right panel).

In the following, the observations made during the summer months during near clear sky conditions are presented. These conditions include situations with a thin and high cirrus cloud cover.
Figure 8: Diurnal cycle of longwave radiative heating rate and the observed temperature tendency or "full" heating rate during summer (near) clear sky days at Summit, Greenland.
Net longwave radiative flux divergence leads to radiative heating/cooling of the same order of magnitude as the observed temperature tendency (Figure. 8). The diurnal variation of longwave radiative flux divergence is shown for 4 layers and the incoming, outgoing, and net flux components in Figures 9 a-d. Close to the surface (0.5 -2 m), the individual components of the longwave radiative heating rates are very large (~ 200 Kd-1). Similar magnitudes have previously been observed by Eliseev et al. (2002). A decrease in the absolute magnitude of longwave radiative heating and cooling is seen with height.
Figure 7: Diurnal cycles of longwave radiative flux divergence (incoming, outgoing and net component) for 4 layers during summer near clear sky conditions at, at Summit, Greenland.

The divergence of the outgoing flux is usually stronger than the divergence of the incoming flux. The divergence of the incoming flux usually opposes the divergence of the outgoing flux. The importance of the divergence of the incoming flux component increases towards the surface.

The characteristic heating rate profile

Figure 8 shows the daytime and nighttime profiles of longwave radiative flux divergence for near clear sky conditions. Close to the surface, daytime cooling and nighttime heating is observed. During the day, a dominating  incoming flux divergence, during the night a dominating convergence of the outgoing flux leads to this result. Above 2 m, the sign changes, a nighttime cooling and a daytime heating results - controlled by the divergence of the outgoing flux. Above 10 m, radiative cooling is observed throughout the entire day.

Figure 8: Profile of the longwave radiative heating rate for near near sky conditions during summer at Summit, Greenland.
A correlation is seen between the temperature gradient within a layer and the net longwave radiative heating rate. In the near surface layer (0.5-2 m) an increased radiative heating is seen with stronger gradients (Figure 9). Above 2 m, the opposite is seen, stronger cooling within increasingly stable layers (Figure 10).
Figure 9 Figure 10

Radiative heating and the fine structure of the temperature profile

During near clear sky daytime conditions during summer, an elevated surface inversion is seen in the temperature profile (Figure 11). A similar feature has previously been observed in Antarctica (Sodemann and Foken, 2004). This inversion lies at the height were a change is observed from strong longwave radiative cooling below to strong radiative heating above (Figure 11). Radiative flux divergence is thus suggested to play an important role in the formation and maintenance of this feature. At nighttime, a characteristic feature is observed in the near surface air layers as well. A layer with a reduced stability between 0.3 m and 5 m is found. It is less stably stratified than the air layers above and below (Figure 12). Again, this pattern is suggested to reflect the variation of longwave radiative flux divergence with height. Near surface radiative heating reduces the stable stratification.



Figure 11
Figure 12

Summary

Longwave radiative flux divergence shows a sign change in the first few meters above the surface. Under stable nighttime (unstable daytime) conditions, a shallow layer of radiative heating (cooling) is observed at the surface. Above about 2 m, radiative cooling (heating) is found. The vertical variation of radiative flux divergence is reflected in the fine structure of the temperature profile near the surface. Nevertheless, the strong evening cooling can not be attributed to radiative flux divergence alone (Fig. 8)  - contrasting the findings of Ha and Mahrt (2003).

References
Eliseev, A. A. et. al.,2002: Izvestiya, Atmospheric and Oceanic Physics, 38 (5), 649-657
Garratt J. R. and  R. A. Brost, 1981: J. Atmos. Sci. 38 (12), 1307-1316
Ha, K. J. and L. Mahrt, 2003: Tellus 55A (4), 317-327
Hoch et al. 2007: J. Appl. Meteorol., accepted
Räisänen P.: 1996: Tellus 48A (3), 403-423
Robinson, G.D.,1950: Q. J. Roy. Meteor. Soc. 76, 37-51
Sodemann H. and T. Foken, 2004: Theor. Appl. Climatol. 80 (2-4), 81-89
Stull, R. B., 1988. Kluver Academic Publishers

IACETH-Gonio-Spectrometer

State of the art instrumentation was developed by C. Saskia Bourgeois to measure spectral and directional reflected solar radiation. The main purpose is the examination of the Hemispherical Directional Reflectance Factor, HDRF, for various snow surfaces and solar zenith angles >> read more.