1. Introduction
Studying the Antarctic atmosphere is difficult because of its challenging climate and sparse observational network, with the lowest near-surface meteorological observation density of any other region. The Antarctic Meteorological Research Center (AMRC) at the University of Wisconsin–Madison has installed dozens of automatic weather stations (AWS) over the past few decades, most of which are nominally 3 m tall (Lazzara et al. 2012b). Most recently, Costanza et al. (2016) studied the climatology of the Ross Ice Shelf using more than a dozen of these stations. Over the past few decades, more stations and better technology have filled in some of the gaps in the Antarctic observational record.
In February 2011, AMRC installed Alexander Tall Tower! (ATT), a 30-m-tall AWS located on the Ross Ice Shelf at 79.023°S, 170.699°E (see Fig. 1). The surface elevation at the tower site is 55 m. There are six observational levels over the 30-m height of the tower that allow for analysis of the lower portion of the boundary layer in this region. Table 1 describes the location of the sensors on the tower, and Fig. 2 illustrates this in schematic form. In total, there are six temperature sensors, four aerovanes, two wind anemometers, two relative humidity sensors, a net shortwave and net longwave radiation sensor, a pressure sensor, and an acoustic depth gauge (to measure snow depth). The heights of each tower level listed in Table 1 are approximate, as snow accumulation and drift can slightly affect these; however, efforts are made to restore these heights when the site is visited every year or two. These approximate heights above ground level (AGL) are used in all computations and figures in this study. The average height of each level over this period of study is slightly different as a result of snow accumulation: levels 1–4 were 0.25 m lower, level 5 was 0.17 m lower, and level 6 was 0.15 m lower. Having multiple levels of observation allows for computation of the sensible and latent heat fluxes using bulk flux formulations with the mean atmospheric measurements from the tower. These flux estimates allow us to quantify this portion of the Ross Ice Shelf’s role as an atmospheric energy sink.
(a) Map showing the location and (b) photograph of ATT on the Ross Ice Shelf in West Antarctica. Both images can be found online (http://amrc.ssec.wisc.edu).
Citation: Journal of Applied Meteorology and Climatology 57, 2; 10.1175/JAMC-D-17-0017.1
(a) Map showing the location and (b) photograph of ATT on the Ross Ice Shelf in West Antarctica. Both images can be found online (http://amrc.ssec.wisc.edu).
Citation: Journal of Applied Meteorology and Climatology 57, 2; 10.1175/JAMC-D-17-0017.1
(a) Map showing the location and (b) photograph of ATT on the Ross Ice Shelf in West Antarctica. Both images can be found online (http://amrc.ssec.wisc.edu).
Citation: Journal of Applied Meteorology and Climatology 57, 2; 10.1175/JAMC-D-17-0017.1
List of observations, the approximate height (m) of each level above the snow surface, instruments, and the manufacturer’s stated accuracy of the instruments. See Fig. 2 for a visual representation of this table. Numbers in parentheses in the approximate heights column are the average height of the instrument over the period studied, February 2011–January 2014.
Schematic depicting the instruments on ATT (J. Snarski helped in creating this figure). Refer to Table 1 for instrument specifics.
Citation: Journal of Applied Meteorology and Climatology 57, 2; 10.1175/JAMC-D-17-0017.1
Schematic depicting the instruments on ATT (J. Snarski helped in creating this figure). Refer to Table 1 for instrument specifics.
Citation: Journal of Applied Meteorology and Climatology 57, 2; 10.1175/JAMC-D-17-0017.1
Schematic depicting the instruments on ATT (J. Snarski helped in creating this figure). Refer to Table 1 for instrument specifics.
Citation: Journal of Applied Meteorology and Climatology 57, 2; 10.1175/JAMC-D-17-0017.1
The purpose of this study is to present the ATT data available via AMRC and present monthly means of meteorological conditions at the tower. This includes temperature, resultant wind speed, relative humidity, pressure, and estimates of sensible and latent heat fluxes. A biannual winter phenomenon is identified in these monthly means, similar to work done by Costanza et al. (2016) using AWS data on the Ross Ice Shelf and Lazzara et al. (2012a) using Amundsen–Scott South Pole Station data. Temperature and pressure data from the West Antarctic Ice Sheet display this interannual variability as well, indicating that it may not be a local phenomenon (Reusch and Alley 2004). Section 2 details the findings of previous studies that have calculated turbulent heat fluxes in the Antarctic. Section 3 describes the data from the tower. Section 4 explains the method for computing heat flux estimates. Section 5 presents the results of monthly averaging and heat flux estimates, and discusses the apparent second initiation of winter earlier in the year. Last, section 6 contains concluding remarks.
2. Background
Computing turbulent heat fluxes with similarity theory has been used in several different regions of Antarctica. These fluxes, while not large in magnitude, can vary based on local environmental conditions, such as specific wind regimes (high plateau vs proximity to the coast). Near-surface measurements across ice shelves, however, are relatively unvarying as a result of the flatness and uniformity of the surface. Consequently, turbulent heat flux estimates would not vary much across the ice shelf either. However, this homogeneity does not necessarily mean that these flux estimates can be ascribed to larger surrounding regions, as wind speeds greatly affect these estimates and change across the ice shelf.
Halley is a research station on the Brunt Ice Shelf of Antarctica, where multiple studies have estimated turbulent fluxes. King (1990) used field campaign measurements from Halley and determined that above 5 m, the temperature and wind profiles cannot be described by similarity theory. Cassano et al. (2001) used tower measurements (and direct eddy covariance calculations) to calculate the fluxes at Halley and compared those with seven model surface layer parameterizations (based on Monin–Obukhov similarity theory). Their results showed that models tend to underestimate the magnitude of the downward sensible heat flux, contrasting the results of even earlier studies by King and Connolley (1997) and Hines et al. (1999).
Van As et al. (2005a) used field campaign measurements to quantify the diurnal cycle of turbulent heat fluxes at Kohnen station in Dronning Maud Land, located on the East Antarctic plateau 500 km from the coast. Other studies in Dronning Maud Land specifically attempt to identify differences in turbulent heat fluxes caused by environmental conditions. van den Broeke et al. (2005) calculated sensible heat flux estimates over a 4-yr period for four AWS in Dronning Maud Land, Antarctica, including two in the katabatic wind zone (one coastal, one inland). The sensible heat flux was found to be highest in katabatic wind zones, especially under clear-sky conditions. A follow-up study by van den Broeke et al. (2006) calculated the diurnal cycle of both sensible and latent heat fluxes at the same four sites during the austral summer and identified a nighttime supply of heat to the surface from the sensible heat flux and a daytime loss of heat from the surface as a result of sublimation (latent heat flux). In the katabatic wind zones, the nocturnal sensible heat flux was enhanced and nighttime radiative heat loss was greatest at the interior AWS (van den Broeke et al. 2004; van den Broeke et al. 2006).
Stearns and Weidner (1993) used the AMRC AWS network to compute monthly mean sensible and latent heat flux estimates. Several stations from different Antarctic regions were studied, including at a site on the Ross Ice Shelf where ATT is located. The monthly mean sensible heat flux estimates were less negative (from atmosphere to surface, the opposite convention used in this study) closer to the Transantarctic Mountains, though the latent heat flux estimates appeared unaffected by this proximity (Stearns and Weidner 1993). Pavolonis et al. (2004) later used these heat flux estimates as validation for a satellite product. The results in the current study are computed using the same methodology as Stearns and Weidner (1993), which makes use of Monin and Obukhov (1954) similarity theory as well as wind speed and temperature profiles as described in Lettau (1979). Sensible and latent heat fluxes are nonnegligible when determining net energy flux in Antarctica. To estimate the surface energy budget, the net shortwave and longwave radiation are also required. Data from the radiometer (on level 1) indicated that the sensor might have tilted as a result of snow accumulation, and thus they could not be utilized in this climatology, so the results below discuss only the turbulent flux contributions to the surface energy budget. Estimates of the latent and sensible heat fluxes are made at the tower site using observations from levels 3, 4, and 6.
3. Data
The instruments at ATT record measurements every 10 s. The data used in this study are from a memory card collected from the tower in January 2014. A few studies have been published using this same data in different manners (Cassano et al. 2016; Wille et al. 2016; Nigro et al. 2017). The raw measurements were processed according to the World Meteorological Organization (WMO) standards of averaging data over time intervals to reduce uncertainties in instantaneous data (WMO 2012). Temperature, relative humidity, and pressure are averaged over a 1-min period, and the wind speed and wind direction are averaged over a 2-min period. The wind speed and direction averages in this study are resultant values. For example, the temperature measurement for 0000 UTC is taken as the mean of data from 2359:10 to 0000:00 UTC, and the wind speed measurement is the resultant of wind speeds from 2358:10 to 0000:00 UTC. This was done at 10-min intervals, which were then quality controlled following procedures outlined in Lazzara et al. (2012b). Data greater than 3 standard deviations from the mean are flagged using a computer program, and data are manually removed when thought to be unrealistic. The relative humidity data presented are with respect to water, the WMO (2012) standard for automatic weather stations. All data presented in the following sections are monthly means from February 2011 to January 2014. Quality-controlled instantaneous scalar wind data were provided by the University of Colorado Boulder. The quality control (QC) was performed for the first two years of available data, February 2011–January 2013, at 10-min intervals. This is compared with the quality-controlled resultant wind in section 5.
An example of the quality-controlled averaged data (available at
Example of the available quality-controlled (a) temperature and (b) wind speed data from AMRC every 10 min. The minimum temperature over the 3-yr period occurred on 21 Sept 2011 at 1 m AGL (vertical black line).
Citation: Journal of Applied Meteorology and Climatology 57, 2; 10.1175/JAMC-D-17-0017.1
Example of the available quality-controlled (a) temperature and (b) wind speed data from AMRC every 10 min. The minimum temperature over the 3-yr period occurred on 21 Sept 2011 at 1 m AGL (vertical black line).
Citation: Journal of Applied Meteorology and Climatology 57, 2; 10.1175/JAMC-D-17-0017.1
Example of the available quality-controlled (a) temperature and (b) wind speed data from AMRC every 10 min. The minimum temperature over the 3-yr period occurred on 21 Sept 2011 at 1 m AGL (vertical black line).
Citation: Journal of Applied Meteorology and Climatology 57, 2; 10.1175/JAMC-D-17-0017.1
4. Methods
To illustrate the advantage of a multilevel tower and to bolster the information gained from several observations per level, the turbulent heat fluxes are computed. Estimating the sensible and latent heat fluxes using similarity theory requires many instruments on ATT. King (1990) found that the surface layer at Halley (on the Brunt Ice Shelf) could be as shallow as 5 m AGL, and thus similarity theory no longer applies to wind and temperature profiles above this height. Surface flux estimates are made using levels 3, 4, and 6 (7.5, 4, and 1 m, respectively) on the tower. The level-3 (or 7.5 m) relative humidity sensor is used with the level-4 temperature and wind sensors to compute the latent heat flux. Level 3 has the lowest (height) relative humidity sensor, level 4 has the lowest wind anemometer, and level 6 is the closest temperature reading to the surface. Using these data reduces the error of surface temperature estimates when using a temperature profile.
a. Profiles for wind speed and temperature
The temperature and wind speed profiles are found using the Monin–Obukhov (Monin and Obukhov 1954) nondimensional height. A more in-depth description of these computed profiles and the stability functions can be found in Lettau (1979). The stability functions from Lettau (1979) were primarily chosen to replicate the work in Stearns and Weidner (1993). There are other vertical profiles that are valid for use in Antarctica and were compared at Dome C, Antarctica, in Vignon et al. (2017). Vignon et al. (2017) found the Lettau (1979) profiles did not markedly underestimate the stability function for θ. The turbulent scales of wind speed (or friction velocity u*) and temperature θ * are approximated using the bulk method.
b. Sensible heat flux estimates
c. Latent heat flux estimates
5. Results and discussion
The three years of meteorological data from Alexander Tall Tower! provide a picture of local conditions and add to the surrounding network of automatic weather station data. The monthly mean of measured temperature, relative humidity, resultant and scalar wind speeds, and pressure are displayed and discussed. Additionally, the absolute humidity is computed and presented as a monthly mean. Last, the sensible and latent heat flux estimates are also presented as monthly means.
a. Temperature
Maximum and minimum measurements over the 3-yr period are listed in Table 2. Figure 4a shows the mean monthly temperatures measured at the tower for the 2011–14 period. The maximum temperature is almost always below freezing, with the most notable exception being the maximum temperature observed at the top level on 27 December 2012 (1.4°C). The minimum temperature recorded was −57.3°C at 1430 UTC 21 September 2011, at level 6, the lowest level of the tower. The maximum temperatures all occur in December, while the minimum temperatures occur during August or September. In the monthly mean there is a persistent inversion year-round (Fig. 4a) as a result of net radiative heat loss, which is characteristic of the Antarctic. Cassano et al. (2016), using ATT data from February 2011 to January 2013, found inversions over the depth of the 30-m tower that exceeded 25°C. The reduced range of temperatures over the depth of the tower during the austral summer, seen in Fig. 4a, is due to decreased net radiative heat loss (more insolation), which weakens the inversion (Phillpot and Zillman 1970), while the winter months have a greater range of temperatures as a result of greater radiative loss at this time of year.
Maximum and minimum temperatures, wind speeds, and pressure over the 3 yr as well as the month in which these measurements were taken. Level 1 is the top of the tower (30 m). Minimum wind speed is not reported, as it is 0.0 m s−1 during calm conditions. Maximum temperature, wind speed, and minimum temperature observed across all levels are highlighted in boldface type.
(a) Monthly mean temperatures at ATT from February 2011 to January 2014. Inversion conditions are observed year-round with the highest temperatures at 30 m. The amount of data missing or dropped during QC from each level is 4.60% (30 m), 4.55% (15 m), 4.69% (7.5 m), 4.98% (4 m), 5.18% (2 m), and 4.88% (1 m). (b) Monthly means of level-1 temperatures for each individual year.
Citation: Journal of Applied Meteorology and Climatology 57, 2; 10.1175/JAMC-D-17-0017.1
(a) Monthly mean temperatures at ATT from February 2011 to January 2014. Inversion conditions are observed year-round with the highest temperatures at 30 m. The amount of data missing or dropped during QC from each level is 4.60% (30 m), 4.55% (15 m), 4.69% (7.5 m), 4.98% (4 m), 5.18% (2 m), and 4.88% (1 m). (b) Monthly means of level-1 temperatures for each individual year.
Citation: Journal of Applied Meteorology and Climatology 57, 2; 10.1175/JAMC-D-17-0017.1
(a) Monthly mean temperatures at ATT from February 2011 to January 2014. Inversion conditions are observed year-round with the highest temperatures at 30 m. The amount of data missing or dropped during QC from each level is 4.60% (30 m), 4.55% (15 m), 4.69% (7.5 m), 4.98% (4 m), 5.18% (2 m), and 4.88% (1 m). (b) Monthly means of level-1 temperatures for each individual year.
Citation: Journal of Applied Meteorology and Climatology 57, 2; 10.1175/JAMC-D-17-0017.1
During the month of May, there is an increase in monthly mean temperature, which is evident in the minute-averaged data as well. Figure 4b illustrates the monthly mean temperature at level 1 (30 m) for each year. The monthly means in May do not vary from the 3-yr mean much, although the 2011 winter is seemingly more muted. This phenomenon in May can be attributed to the circumpolar trough being closer to the pole during the autumn months, which slows or reverses the progression of the winter season (van Loon 1967; van den Broeke 2000). The break in the winter season as a result of this circulation anomaly is also apparent in wind speeds and pressure. Lazzara et al. (2012a) and Costanza et al. (2016) referred to the two decreases of temperature (seen in April and June surrounding the May temperature increase) as two separate winter seasons. Here, it is referred to as two winter regimes.
b. Humidity
Monthly mean relative humidity measured at levels 1 and 3 are plotted in Fig. 5a. Monthly mean absolute humidity is plotted in Fig. 5b. On average, relative humidity is always higher at lower heights on the tower, where the temperatures are lower. Relative humidity is highest in the austral summer and lowest in the winter. However, it is an important caveat that the relative humidity data illustrated here and available through AMRC are measured with respect to water. There is a local maximum in the relative humidity in May, when, thus far, there has been a local maximum in temperature and wind speeds and a local minimum in pressures for 2011 and 2012, possibly as a result of the aforementioned early winter synoptic anomaly. The absolute humidity, shown in Fig. 5b, is computed using the relative humidity, the ideal gas law, and Eq. (11). The absolute humidity is highest in the austral summer, with a local maximum in monthly averages during the month of May. The discrepancies between levels 1 and 3 are minimal, indicating that vertically atmospheric moisture content does not vary much. This is justification for using the relative humidity at level 3 (7.5 m) to compute the sensible and latent heat flux estimates.
Monthly mean (a) relative humidity and (b) absolute humidity at ATT from February 2011 to January 2014. The amount of data missing or dropped during QC from each level is 4.64% (30 m) and 4.36% (7.5 m).
Citation: Journal of Applied Meteorology and Climatology 57, 2; 10.1175/JAMC-D-17-0017.1
Monthly mean (a) relative humidity and (b) absolute humidity at ATT from February 2011 to January 2014. The amount of data missing or dropped during QC from each level is 4.64% (30 m) and 4.36% (7.5 m).
Citation: Journal of Applied Meteorology and Climatology 57, 2; 10.1175/JAMC-D-17-0017.1
Monthly mean (a) relative humidity and (b) absolute humidity at ATT from February 2011 to January 2014. The amount of data missing or dropped during QC from each level is 4.64% (30 m) and 4.36% (7.5 m).
Citation: Journal of Applied Meteorology and Climatology 57, 2; 10.1175/JAMC-D-17-0017.1
c. Winds
Monthly resultant wind speeds are shown in Fig. 6a and monthly scalar wind speeds in Fig. 6b. The maximum resultant wind speeds are given in Table 2. The level-5 (2 m) wind speeds were not included, as they were measured with a cup anemometer and there is no wind direction observed at that level. Figure 6a shows that the resultant wind speed over the 30-m depth of observations increases with altitude, as expected from similarity theory, and increases sharply in February and decreases sharply in December. The maximum resultant (2 min) wind speeds all occur at the same time (0250 UTC) on 9 August 2013, indicating a strong wind event that extended to at least 30 m in the boundary layer. In Fig. 6a, the maximum monthly resultant wind speeds for all four levels occur in May, which was previously noted to have an increase in monthly averaged temperatures. Again, this indicates the pause in the winter season.
Monthly (a) resultant wind speeds (from February 2011 to January 2014) and (b) scalar wind speeds (from February 2011 to January 2013) at ATT. Wind speeds increase with height and are weakest during the austral summer (December–February). The amount of data missing or dropped during QC of the resultant wind from each level is 4.17% (30 m), 4.73% (15 m), 5.08% (7.5 m), and 7.60% (4 m). Scalar wind data were not quality controlled by the AMRC.
Citation: Journal of Applied Meteorology and Climatology 57, 2; 10.1175/JAMC-D-17-0017.1
Monthly (a) resultant wind speeds (from February 2011 to January 2014) and (b) scalar wind speeds (from February 2011 to January 2013) at ATT. Wind speeds increase with height and are weakest during the austral summer (December–February). The amount of data missing or dropped during QC of the resultant wind from each level is 4.17% (30 m), 4.73% (15 m), 5.08% (7.5 m), and 7.60% (4 m). Scalar wind data were not quality controlled by the AMRC.
Citation: Journal of Applied Meteorology and Climatology 57, 2; 10.1175/JAMC-D-17-0017.1
Monthly (a) resultant wind speeds (from February 2011 to January 2014) and (b) scalar wind speeds (from February 2011 to January 2013) at ATT. Wind speeds increase with height and are weakest during the austral summer (December–February). The amount of data missing or dropped during QC of the resultant wind from each level is 4.17% (30 m), 4.73% (15 m), 5.08% (7.5 m), and 7.60% (4 m). Scalar wind data were not quality controlled by the AMRC.
Citation: Journal of Applied Meteorology and Climatology 57, 2; 10.1175/JAMC-D-17-0017.1
The monthly mean of the scalar wind, Fig. 6b, consists of one less year than the resultant wind data and stops in January 2013. The scalar wind appears to be consistently higher than the resultant wind speeds throughout the year, but it shows an annual trend of increased wind speeds during the austral winter and decreased wind speeds in the austral summer. The constancy, defined as the ratio of the scalar wind to the resultant wind, was computed for each level at each time step. The average constancy over the two years for each level was 83% at level 1, 84% at levels 2 and 3, and 90% at level 4. The constancy increases with decreasing altitude, indicating more directionally consistent winds at lower levels. Annually, the constancy is highest during the austral winter for all levels and lowest during the austral summer.
d. Pressure
Figure 7 shows the monthly mean pressures measured at approximately 3 m above ground. Instead of averaging all months together, as the interannual variability is large, it is plotted as a time series. The month of June exhibits a local maximum in pressure, with especially higher mean pressure in 2012 and 2013, possibly indicative of the end of the early winter pause. Both the maximum and minimum pressure were recorded in August 2011, with a pressure of 1020.3 hPa at 1630 and 1640 UTC 2 August and 945.8 hPa from 0940 to 1000 UTC 23 August, respectively. It is important to denote that these are 1-min averages of pressures and not the 10-s instantaneous values. There is a sharp increase in average pressure from May to June each year, and a sharp decrease between June and July. This same pattern marking the two winter regimes was found at South Pole Station (Lazzara et al. 2012a) and is again caused by the circumpolar trough moving toward the pole in autumn (van Loon 1967; van den Broeke 2000).
Monthly mean pressure at ATT from February 2011 to January 2014. The amount of data missing or dropped during QC is 4.11%.
Citation: Journal of Applied Meteorology and Climatology 57, 2; 10.1175/JAMC-D-17-0017.1
Monthly mean pressure at ATT from February 2011 to January 2014. The amount of data missing or dropped during QC is 4.11%.
Citation: Journal of Applied Meteorology and Climatology 57, 2; 10.1175/JAMC-D-17-0017.1
Monthly mean pressure at ATT from February 2011 to January 2014. The amount of data missing or dropped during QC is 4.11%.
Citation: Journal of Applied Meteorology and Climatology 57, 2; 10.1175/JAMC-D-17-0017.1
e. Heat fluxes
For the heat flux estimates, the assumptions made are similar to those made in Stearns and Weidner (1993): eddy diffusivities for momentum and moisture are assumed to be equal, and the latent heat flux is zero when the wind speed is zero. If inversion conditions cannot be determined as a result of QC removal of data from one or both levels, then the heat flux estimate is not made. When the surface temperature is computed [using Eq. (2)] to be above 0°C, this is assumed to be physically unrealistic, as a snow surface cannot have a temperature above the melting point, and no heat flux is calculated. A positive surface temperature was computed fewer than 1600 times, which is approximately 0.977% of the data. This corresponds to 10.7 days of data removed, which was found to occur during the austral summer. When the level-4 wind speed is less than 3 m s−1 and the vertical air temperature difference is larger than 0°C (inversion conditions), the computation for sensible heat flux will not always converge to a solution. In this case, the sensible and latent heat fluxes are not computed, approximately 16.8% of the time. As previously reported, data from January 2013 were removed from these estimates, as there were many missing data points from the QC of data, which caused biases in the monthly mean.
f. Sensible heat flux
The sensible heat flux estimates are plotted in Fig. 8, with positive values indicating a downward flux from the atmosphere to the surface. An annual pattern is identified with the heat flux close to 0 W m−2, although still downward, in the summer months (December and January) and a maximum downward flux in the winter months (June–August). This annual shape inversely matches that of the temperatures in Fig. 4a, as the lower sensible heat flux estimates during austral summer are caused by more insolation. The larger downward flux in winter is also consistent with the stronger inversion conditions seen in Fig. 4a.
As in Fig. 7, but for sensible heat flux estimates. The number of missing data points is due to the QC of variables used to compute these conditions, unavailable data, or not meeting criteria for an estimate to be computed (see text). This is 26.8% of the total timeline.
Citation: Journal of Applied Meteorology and Climatology 57, 2; 10.1175/JAMC-D-17-0017.1
As in Fig. 7, but for sensible heat flux estimates. The number of missing data points is due to the QC of variables used to compute these conditions, unavailable data, or not meeting criteria for an estimate to be computed (see text). This is 26.8% of the total timeline.
Citation: Journal of Applied Meteorology and Climatology 57, 2; 10.1175/JAMC-D-17-0017.1
As in Fig. 7, but for sensible heat flux estimates. The number of missing data points is due to the QC of variables used to compute these conditions, unavailable data, or not meeting criteria for an estimate to be computed (see text). This is 26.8% of the total timeline.
Citation: Journal of Applied Meteorology and Climatology 57, 2; 10.1175/JAMC-D-17-0017.1
The closest AWS to ATT that Stearns and Weidner (1993) computed heat flux estimates for was Schwerdtfeger, during the period 1985–90 (excluding 1987). The monthly sensible heat flux estimates at Schwerdtfeger were between −42 and −2 W m−2 (opposite sign convention as this study). By comparison, the ATT monthly sensible heat flux estimates were within 1 and 39 W m−2. The annual mean sensible heat flux estimates at ATT, for the three years of this study, are between approximately 19 and 23 W m−2. van den Broeke et al. (2005) calculated annual mean sensible heat fluxes in katabatic wind zones as being 22–24 W m−2 toward the surface. Van As et al. (2005a) found a daily mean sensible heat flux of 8 W m−2 (toward the surface) in Dronning Maud Land, a katabatic wind zone, in January and February 2002. These studies agree that on average in these katabatic wind zones, the sensible heat flux is toward the surface, just as was computed for ATT.
The observed inversion conditions agree with a sensible heat flux toward the surface, as the near-surface air is cooled by this flux. In January and December, where the temperatures are highest annually, the 1-min averaged sensible heat flux estimates are the smallest magnitude, which is reflected in the smaller temperature gradient during these months—the atmosphere is well mixed in these conditions. The greater-magnitude sensible heat flux estimates during nonsummer months are caused by large vertical temperature gradients and high wind speeds, which Stearns and Weidner (1993) attribute to katabatic flow from Byrd Glacier. Several satellite analyses support that katabatic flow originating from Byrd Glacier as well as Marie Byrd Land propagates through the vicinity of tall tower during nonsummer months (Bromwich 1989a,b; Bromwich et al. 1992).
g. Latent heat flux
The latent heat flux estimates are plotted in Fig. 9, and the monthly means are much smaller in magnitude in comparison with the sensible heat flux estimates for every season other than austral summer. The annual mean latent heat flux estimates are between −2 and −4 W m−2. In December, the latent heat flux estimates are greatest in magnitude, like the sensible heat flux estimates at this time of year. The larger magnitude of the upward moisture flux during the austral summer increases the absolute humidity of the air, as evidenced in Fig. 5b.
As in Fig. 8, but for latent heat flux estimates. Missing data points are 26.4% of the total timeline.
Citation: Journal of Applied Meteorology and Climatology 57, 2; 10.1175/JAMC-D-17-0017.1
As in Fig. 8, but for latent heat flux estimates. Missing data points are 26.4% of the total timeline.
Citation: Journal of Applied Meteorology and Climatology 57, 2; 10.1175/JAMC-D-17-0017.1
As in Fig. 8, but for latent heat flux estimates. Missing data points are 26.4% of the total timeline.
Citation: Journal of Applied Meteorology and Climatology 57, 2; 10.1175/JAMC-D-17-0017.1
Stearns and Weidner (1993) computed monthly latent heat flux estimates of between −1 and 11 W m−2 at Schwerdtfeger site (again, the opposite sign convention), whereas the ATT monthly latent heat flux estimates are between −8 and 0 W m−2. Van As et al. (2005b) computed the daily latent heat flux to range from −4 to 0 W m−2 during January and February 2002 in Dronning Maud Land, which is in a katabatic wind zone. There is agreement that averaging the latent heat flux over longer periods results in sublimation over deposition. Before averaging, the maximum amplitude of the latent heat flux at Dronning Maud Land is 2 W m−2 (van As et al. 2005a); in this study, during the same period, the maximum amplitude is 10 W m−2. In both cases the maximum amplitude is directed to the surface (deposition).
h. Flux discussion
Both the monthly mean sensible and latent heat flux estimates have an annual cycle, though of opposite sign. The sensible heat flux, on average, is directed toward the surface throughout the year with the largest flux occurring during winter months. The latent heat flux, on average, is directed toward the atmosphere throughout the year, with the largest flux occurring during the summer months. The positive heat flux estimates indicate a sink of atmospheric energy in the Antarctic, while the negative heat flux estimates would indicate a source of energy. These computed heat flux estimates are comparable in both magnitude and direction to those computed in Antarctic katabatic wind zones (Stearns and Weidner 1993; van den Broeke et al. 2005; van As et al. 2005a,b).
To further test the estimation of the turbulent heat fluxes, Table 3 shows the annual range of monthly mean sensible and latent heat fluxes using four different stability correction functions. For stable conditions the correction functions used are from Holtslag and De Bruin (1988), Grachev et al. (2007), King and Anderson (1994), and Högström (1996). For unstable cases, the stability functions from Högström (1996) were used; this combination of stable and unstable functions was done in Vignon et al. (2017). The annual range of turbulent heat fluxes varies little, with the largest difference being the lower sensible heat flux maximum when using the Högström (1996) stability correction functions. Future studies, which will include more tall towers (exceeding the standard 3-m height) in Antarctica, will further investigate the turbulent heat fluxes and the manner in which they can be estimated in these remote locations.
Annual range of monthly mean sensible and latent heat flux estimates (W m−2) based on the stability functions used. For Holtslag and De Bruin (1988), Grachev et al. (2007), and King and Anderson (1994), the Högström (1996) unstable stability functions were used, as was done in Vignon et al. (2017).
6. Conclusions
Alexander Tall Tower! has had a relatively short lifetime as an Antarctic AWS, but it is providing consistent and beneficial data. Observations of meteorological conditions at the six tower levels provide details of the lower boundary layer in this otherwise unobserved region. The presence of two winter regimes annually, caused by circulation anomalies, is evident in these observations. Using the multiple levels of instrumentation, sensible and latent heat flux estimates were calculated over three years. Positive sensible heat flux estimates ranging from 1 to 39 W m−2 (monthly mean) indicate a cooling atmosphere, and negative latent heat flux estimates ranging from −8 to 0 W m−2 indicate sublimation of snow. The latent heat flux estimates are smaller in magnitude when compared with the sensible heat flux estimates, though both have an apparent annual cycle. The positive sensible heat flux estimates are large as a result of large temperature gradients and higher wind speeds forced by katabatic flow and support the strong inversion conditions noted in the monthly mean temperatures. Similar magnitude and direction of these fluxes were found using multiple different stability correction functions. Future estimations of turbulent fluxes will further investigate different models of stability and will use measurements from multiple tall towers.
These heat fluxes are an important component of the atmospheric energy budget, given that Antarctica acts as a global heat sink. To produce the most accurate estimates, the conditions of the tower itself must be well known. Observational data are prone to error when instruments are damaged or the tower is tilted or buried. Installing more tall towers in the Antarctic would allow for more heat flux estimates to be computed and would create a broader picture of the energy budget in this region. Future work includes comparisons between turbulent heat flux estimates from multiple tall towers, as well as comparisons with the Antarctic Mesoscale Prediction System (AMPS) fluxes, and incorporation of radiation flux data to perform a surface energy balance analysis. In addition, more towers could confirm the dual-winter regime phenomenon that is unique to the Antarctic. This could affect travel in this region and thus is an important feature to document and understand. Greater temporal and spatial coverage would allow further analysis of the consequences of this phenomenon in Antarctic weather and climate.
Acknowledgments
The authors appreciate the support of the Office of Polar Programs at the National Science Foundation (Grants ANT-0944018, ANT-1141908, ANT-1245663, ANT-1543305, OPP-2345737, and OPP-1543158) for their support of the U.S. Antarctic AWS program and the Antarctic Meteorological Research Center. This material is based on (data, equipment, and/or engineering) services provided by the UNAVCO Facility with support from the National Science Foundation (NSF) and the National Aeronautics and Space Administration (NASA) under NSF Cooperative Agreement EAR-0735156. Thanks are given to Melissa Nigro of the University of Colorado Boulder for providing quality-controlled wind data and to Joey Snarski for his help in creating the depiction of the tower and instrumentation. Last, we thank Ankur Desai of the University of Wisconsin–Madison for guidance with the turbulent heat flux estimates.
REFERENCES
Amory, C., and Coauthors, 2015: Comparison between observed and simulated aeolian snow mass fluxes in Adélie Land, East Antarctica. Cryosphere, 9, 1373–1383, https://doi.org/10.5194/tc-9-1373-2015.
Bromwich, D. H., 1989a: Satellite analyses of Antarctic katabatic wind behavior. Bull. Amer. Meteor. Soc., 70, 738–749, https://doi.org/10.1175/1520-0477(1989)070<0738:SAOAKW>2.0.CO;2.
Bromwich, D. H., 1989b: Satellite observations of katabatic winds blowing from Marie Byrd Land onto the Ross Ice Shelf. Antarct. J. U.S., 24, 218–221.
Bromwich, D. H., J. F. Carrasco, and C. R. Stearns, 1992: Satellite observations of katabatic-wind propagation for great distances across the Ross Ice Shelf. Mon. Wea. Rev., 120, 1940–1949, https://doi.org/10.1175/1520-0493(1992)120<1940:SOOKWP>2.0.CO;2.
Cassano, J. J., T. R. Parish, and J. C. King, 2001: Evaluation of surface flux parameterizations for the stable surface layer over Halley, Antarctica. Mon. Wea. Rev., 129, 26–46, https://doi.org/10.1175/1520-0493(2001)129<0026:EOTSFP>2.0.CO;2.
Cassano, J. J., M. A. Nigro, and M. A. Lazzara, 2016: Characteristics of the near-surface atmosphere over the Ross Ice Shelf, Antarctica. J. Geophys. Res. Atmos., 121, 3339–3362, https://doi.org/10.1002/2015JD024383.
Costanza, C. A., M. A. Lazzara, L. M. Keller, and J. J. Cassano, 2016: The surface climatology of the Ross Ice Shelf Antarctica. Int. J. Climatol., 36, 4929–4941, https://doi.org/10.1002/joc.4681.
Grachev, A. A., E. L Andreas, C. W. Fairall, P. S. Guest, and P. O. G. Persson, 2007: SHEBA flux–profile relationships in the stable atmospheric boundary layer. Bound.-Layer Meteor., 124, 315–333, https://doi.org/10.1007/s10546-007-9177-6.
Hines, K. M., R. W. Grumbine, D. H. Bromwich, and R. I. Cullather, 1999: Surface energy balance of the NCEP MRF and NCEP–NCAR reanalysis in Antarctic latitudes during FROST. Wea. Forecasting, 14, 851–866, https://doi.org/10.1175/1520-0434(1999)014<0851:SEBOTN>2.0.CO;2.
Högström, U., 1996: Review of some basic characteristics of the atmospheric surface layer. Bound.-Layer Meteor., 78, 215–246, https://doi.org/10.1007/BF00120937.
Holtslag, A. A. M., and H. A. R. De Bruin, 1988: Applied modeling of the nighttime surface energy balance over land. J. Appl. Meteor., 27, 689–704, https://doi.org/10.1175/1520-0450(1988)027<0689:AMOTNS>2.0.CO;2.
King, J. C., 1990: Some measurements of turbulence over an Antarctic ice shelf. Quart. J. Roy. Meteor. Soc., 116B, 379–400, https://doi.org/10.1002/qj.49711649208.
King, J. C., and P. S. Anderson, 1994: Heat and water vapour fluxes and scalar roughness lengths over an Antarctic ice shelf. Bound.-Layer Meteor., 69, 101–121, https://doi.org/10.1007/BF00713297.
King, J. C., and W. M. Connolley, 1997: Validation of the surface energy balance over the Antarctic ice sheets in the U.K. Meteorological Office Unified Climate Model. J. Climate, 10, 1273–1287, https://doi.org/10.1175/1520-0442(1997)010<1273:VOTSEB>2.0.CO;2.
Lazzara, M. A., L. M. Keller, T. Markle, and J. Gallagher, 2012a: Fifty-year Amundsen–Scott South Pole station surface climatology. Atmos. Res., 118, 240–259, https://doi.org/10.1016/j.atmosres.2012.06.027.
Lazzara, M. A., G. A. Weidner, L. M. Keller, J. E. Thom, and J. J. Cassano, 2012b: Antarctic Automatic Weather Station Program: 30 years of polar observations. Bull. Amer. Meteor. Soc., 93, 1519–1537, https://doi.org/10.1175/BAMS-D-11-00015.1.
Lettau, H. H., 1979: Wind and temperature profile prediction for diabatic surface layers including strong inversion cases. Bound.-Layer Meteor., 17, 443–464, https://doi.org/10.1007/BF00118610.
Monin, A. S., and A. M. Obukhov, 1954: Basic laws of turbulent mixing in the atmosphere near the ground. Tr. Geofiz. Inst., Akad. Nauk SSSR, 24, 163–187.
Nigro, M. A., and J. J. Cassano, 2014: Identification of surface wind patterns over the Ross Ice Shelf, Antarctica, using self-organizing maps. Mon. Wea. Rev., 142, 2361–2378, https://doi.org/10.1175/MWR-D-13-00382.1.
Nigro, M. A., J. J. Cassano, J. Wille, D. H. Bromwich, and M. A. Lazzara, 2017: A self-organizing-map-based evaluation of the Antarctic Mesoscale Prediction System using observations from a 30-m instrumented tower on the Ross Ice Shelf, Antarctica. Wea. Forecasting, 32, 223–242, https://doi.org/10.1175/WAF-D-16-0084.1.
Parish, T. R., J. J. Cassano, and M. W. Seefeldt, 2006: Characteristics of the Ross Ice Shelf air stream as depicted in Antarctic Mesoscale Prediction System simulations. J. Geophys. Res., 111, D12109, https://doi.org/10.1029/2005JD006185.
Pavolonis, M. J., J. R. Key, and J. J. Cassano, 2004: A study of the Antarctic surface energy budget using a polar regional atmospheric model forced with satellite-derived cloud properties. Mon. Wea. Rev., 132, 654–661, https://doi.org/10.1175/1520-0493(2004)132<0654:ASOTAS>2.0.CO;2.
Phillpot, H. R., and J. W. Zillman, 1970: The surface temperature inversion over the Antarctic continent. J. Geophys. Res., 75, 4161–4169, https://doi.org/10.1029/JC075i021p04161.
Reusch, D. B., and R. B. Alley, 2004: A 15-year West Antarctic climatology from six automatic weather station temperature and pressure records. J. Geophys. Res., 109, D04103, https://doi.org/10.1029/2003JD004178.
Stearns, C. R., and G. A. Weidner, 1993: Sensible and latent heat flux estimates in Antarctica. Antarctic Meteorology and Climatology: Studies Based on Automatic Weather Stations, D. H. Bromwich and C. R. Stearns, Eds., Antarctic Research Series, Vol. 61, Amer. Geophys. Union, 109–138.
van As, D., M. R. van den Broeke, and R. S. W. van de Wal, 2005a: Daily cycle of the surface layer and energy balance on the high Antarctic Plateau. Antarct. Sci., 17, 121–133, https://doi.org/10.1017/S095410200500252X.
van As, D., M. R. van den Broeke, C. H. Reijmer, and R. S. W. van de Wal, 2005b: The summer surface energy balance of the high Antarctic Plateau. Bound.-Layer Meteor., 115, 289–317, https://doi.org/10.1007/s10546-004-4631-1.
van den Broeke, M. R., 2000: On the interpretation of Antarctic temperature trends. J. Climate, 13, 3885–3889, https://doi.org/10.1175/1520-0442(2000)013<3885:OTIOAT>2.0.CO;2.
van den Broeke, M. R., C. H. Reijmer, and R. S. W. van de Wal, 2004: Surface radiation balance in Antarctica as measured with automatic weather stations. J. Geophys. Res., 109, D09103, https://doi.org/10.1029/2003JD004394.
van den Broeke, M. R., D. van As, C. H. Reijmer, and R. S. W. van de Wal, 2005: Sensible heat exchange at the Antarctic snow surface: A study with automatic weather stations. Int. J. Climatol., 25, 1081–1101, https://doi.org/10.1002/joc.1152.
van den Broeke, M. R., C. H. Reijmer, D. van As, and W. Boot, 2006: Daily cycle of the surface energy balance in Antarctica and the influence of clouds. Int. J. Climatol., 26, 1587–1605, https://doi.org/10.1002/joc.1323.
van Loon, H., 1967: The half-yearly oscillations in middle and high southern latitudes and the coreless winter. J. Atmos. Sci., 24, 472–486, https://doi.org/10.1175/1520-0469(1967)024<0472:THYOIM>2.0.CO;2.
Vignon, E., C. Genthon, H. Barral, C. Amory, G. Picard, H. Gallée, G. Casasanta, and S. Argentini, 2017: Momentum- and heat-flux parametrization at Dome C, Antarctica: A sensitivity study. Bound.-Layer Meteor., 162, 341–367, https://doi.org/10.1007/s10546-016-0192-3.
Wille, J. D., D. H. Bromwich, M. A. Nigro, J. J. Cassano, M. Mateling, M. A. Lazzara, and S.-H. Wang, 2016: Evaluation of the AMPS boundary layer simulations on the Ross Ice Shelf with tower observations. J. Appl. Meteor. Climatol., 55, 2349–2367, https://doi.org/10.1175/JAMC-D-16-0032.1.
WMO, 2012: Measurements at automatic weather stations. Guide to Meteorological Instruments and Methods of Observation, 8th ed., WMO-8, 526–551, https://library.wmo.int/opac/doc_num.php?explnum_id=3179.
