• Ahrens, C. D., 1994: Meteorology Today. West Publishing, 591 pp.

  • Baker, D. G., R. H. Skaggs, and D. L. Ruschy, 1991: Snow depth required to mask the underlying surface. J. Appl. Meteor.,30, 387–392.

  • ——, D. L. Ruschy, R. H. Skaggs, and D. B. Wall, 1992: Air temperature and radiation depression associated with a snow cover. J. Appl. Meteor.,31, 247–254.

  • Barnett, T. P., L. Dumenil, U. Schlese, E. Roeckner, and M. Latif, 1989: The effect of Eurasian snow cover on regional and global climate variations. J. Atmos. Sci.,46, 661–685.

  • Cerveny, R. S., and R. C. Balling, 1992: The impact of snow cover on diurnal temperature readings. Geophys. Res. Lett.,19, 797–800.

  • Dewey, K. F., 1977: Daily maximum and minimum temperature forecasts and the influence of snow cover. Mon. Wea. Rev.,105, 1594–1597.

  • ——, 1987: Snow cover–atmosphere interactions. Large-Scale Effects of Seasonal Snow Cover: Proceedings of the Vancouver Symposium, Int. Assoc. Hydrol. Sci. Publ. 166, 27–42.

  • Dey, B., and B. Kumar, 1983: Himalayan winter snow cover area and summer monsoon rainfall over India. J. Geophys. Res.,88, 5471–5474.

  • Dickson, R. R., 1984: Eurasian snow cover versus Indian monsoon rainfall—An extension of the Hahn–Shukla results. J. Climate Appl. Meteor.,23, 171–173.

  • Ellis, A. W., and D. J. Leathers, 1998: A quantitative approach to evaluating the effects of snow cover on cold air mass temperatures across the U.S. Great Plains. Wea. Forecasting,13, 688–701.

  • Gutzler, D. S., and R. D. Rosen, 1992: Interannual variability of wintertime snow cover across the northern hemisphere. J. Climate,5, 1441–1447.

  • Heim, R., Jr., and K. F. Dewey, 1984: Circulation patterns and temperature fields associated with extensive snow cover on the North American continent. Phys. Geogr.,4, 66–85.

  • Idso, S. B., 1981: A set of equations for full spectrum and 8–14 μm and 10.5–12.5 μm thermal radiation from cloudless skies. Water Resour. Res.,17, 295–304.

  • Jordan, R., 1991: A one dimensional temperature model for a snow cover: Technical documentation for SNTHERM.89. Special Rep. 91-16, U.S. Army Corps of Engineers, Cold Regions Research and Engineering Laboratory, Hanover, NH. [Available from U.S. Army Corps of Engineers, Cold Regions Research and Engineering Laboratory, Hanover, NH 03755-1290.].

  • Lamb, H. H., 1950: Types and spells of weather around the year in the British Isles: Annual trends, seasonal structure of the year, singularities. Quart. J. Roy. Meteor. Soc.,76, 393–438.

  • Leathers, D. J., and D. A. Robinson, 1993: The association between extremes in North American snow cover extent and United States temperatures. J. Climate,6, 1345–1355.

  • ——, A. W. Ellis, and D. A. Robinson, 1995: Characteristics of temperature depressions associated with snow cover across the northeast United States. J. Appl. Meteor.,34, 381–390.

  • Namias, J., 1962: Influences of abnormal surface heat sources and sinks on Atmospheric behavior. Proceedings of the International Symposium on Numerical Weather Prediction, 1960, Meteorological Society of Japan, 615–627.

  • ——, 1978: Multiple causes of the North American abnormal winter 1976–77. Mon. Wea. Rev.,106, 279–295.

  • ——, 1985: Some empirical evidence for the influence of snow cover on temperature and precipitation. Mon. Wea. Rev.,113, 1542–1553.

  • Robinson, D. A., 1993: Historical daily climatic data for the United States. Preprints, Eighth Conf. on Applied Climatology, Anaheim, CA, Amer. Meteor. Soc., 264–269.

  • ——, A. Frei, D. J. Leathers, and M. C. Serreze, 1995a: Northern Hemisphere snow cover during the transition seasons. Proc. 19th Annual Climate Diagnostics Workshop, College Park, MD, NOAA, 377–380.

  • ——, ——, and M. C. Serreze, 1995b: Recent variations and regional relationships in northern hemisphere snow cover. Ann. Glaciol.,21, 71–76.

  • Ross, B., and J. E. Walsh, 1986: Synoptic scale influence of snow cover and sea ice. Mon. Wea. Rev.,114, 1795–1810.

  • Shapiro, R., 1987: A simple model for the calculation of the flux of direct and diffuse solar radiation through the atmosphere. ST Systems Corporation Scientific Rep. 35, Lexington, MA, 49 pp. [Available from Rachel Jordan, Cold Regions Research Engineering Laboratory, Hanover, NH 03755-1290.].

  • Walsh, J. E., D. R. Tucek, and M. R. Peterson, 1982: Seasonal snow cover and short-term climatic fluctuations over the United States. Mon. Wea. Rev.,110, 1474–1485.

  • ——, W. H. Jasperson, and B. Ross, 1985: Influence of snow cover and soil moisture on monthly air temperature. Mon. Wea. Rev.,113, 756–768.

  • Willmott, C. J., 1984: On the evaluation of model performance in physical geography. Spatial Statistics and Models, G. L. Gaile and C. J. Willmott, Eds., D. Reidel, 443–460.

  • View in gallery
    Fig. 1.

    The study region encompassing the U.S. Great Plains.

  • View in gallery
    Fig. 2.

    The station distributions for (a) snow depth and (b) meteorological data.

  • View in gallery
    Fig. 3.

    The movement of airmass cores one through four (a)–(d), with the positions based on an interval of 12 h. The snow cover extent (≥2.5 cm) for the time period associated with each cold airmass core is also indicated with a heavy line.

  • View in gallery
    Fig. 4.

    Mean simulated and measured average airmass core temperatures (°C) through the time periods of simulation for cold airmass scenarios one through four (a)–(d).

  • View in gallery
    Fig. 5.

    Mean simulated airmass core temperatures (°C) produced by snow albedo conditions of 0.90, 0.80, and 0.50 through the time periods of simulation for cold airmass scenarios one through four (a)–(d).

  • View in gallery
    Fig. 6.

    Mean simulated airmass core sensible heat flux values (W m−2) produced by snow albedo conditions of 0.90, 0.80, and 0.50 through the time periods of simulation for cold airmass scenarios one through four (a)–(d).

  • View in gallery
    Fig. 7.

    Same as Fig. 6 but for latent heat flux.

  • View in gallery
    Fig. 8.

    Same as Fig. 6 but for net longwave radiation.

  • View in gallery
    Fig. 9.

    The spatial distribution of the difference between the mean daytime airmass core temperatures (°C) associated with snow albedo conditions of 0.50 and 0.90 throughout the full simulation periods for scenarios one through four (a)–(d). Differences are taken as temperatures associated with an albedo of 0.50 minus temperatures associated with an albedo of 0.90.

  • View in gallery
    Fig. 10.

    (a) Mean simulated airmass core temperatures (°C), (b) sensible heat fluxes (W m−2), (c) latent heat fluxes (W m−2), and (d) net longwave radiation values (W m−2) produced by snow depth conditions of 2.5, 15.0, and 30.0 cm through the time period of simulation for cold airmass scenario one.

  • View in gallery
    Fig. 11.

    Mean simulated ground heat flux values (W m−2) produced by snow depth conditions of 2.5, 15.0, and 30.0 cm through the time periods of simulation for cold airmass scenarios one through four (a)–(d).

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 220 45 4
PDF Downloads 84 43 6

Analysis of Cold Airmass Temperature Modification across the U.S. Great Plains as a Consequence of Snow Depth and Albedo

Andrew W. EllisDepartment of Geography, Arizona State University, Tempe, Arizona

Search for other papers by Andrew W. Ellis in
Current site
Google Scholar
PubMed
Close
and
Daniel J. LeathersCenter for Climatic Research, Department of Geography, University of Delaware, Newark, Delaware

Search for other papers by Daniel J. Leathers in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

The presence of snow cover has been shown to modify atmospheric conditions through much of the earth’s troposphere due to its radiative effects. Snow cover has garnered much attention in recent decades as a result of concerns associated with potential changes in the global environment that may be intensified by the presence or absence of a snow cover. As a result, a greater emphasis has been placed on the representation of snow cover in weather and climate prediction models. This study investigates the effects of snow albedo and snow depth on the modification of surface air temperatures within cold air masses moving across the U.S. Great Plains in winter.

Through the adaptation of a one-dimensional snowpack model, the thermal characteristics of the core of a cold air mass were derived from the equation governing the heat balance between the surface and the lower atmosphere. The methodology was based on the premise that the core of a cold air mass may be considered homogeneous and not subject to advection of air from outside, thereby isolating the exchange of energy between the surface and the atmosphere as the control on lower-tropospheric temperatures. The adapted model included the synergism of the air mass–snow cover relationship through time, incorporating the natural feedback process.

Simulation of surface air temperatures within four cold air masses over snow cover of different albedo values and depths led to several conclusions. In testing the effects of snow albedo, results indicate 1) mean daytime air temperatures 3°–6°C higher and maximum daytime air temperatures 7°–12°C higher over snow with an albedo equal to 0.50 compared to 0.90, as a consequence of differences in sensible heat flux, and ultimately, absorbed solar radiation, and 2) little thermal inertia and therefore little difference in subsequent nighttime airmass temperatures over snow with an albedo of 0.50 compared to 0.90. In testing the effects of snow depth, results indicate 1) little difference in daytime air temperatures associated with a snow depth of 2.5 cm compared to 15.0 or 30.0 cm, 2) an increase in mean nighttime temperatures of 0.2°–0.7°C over a snow depth of 2.5 cm compared to either of the larger depths, and 3) a masking of the underlying bare soil surfaces by the snow depths of 15.0 and 30.0 cm and virtually no difference in airmass temperatures over the two snow depths.

The potential utility of the results of this study lies in their application as additional guidance for temperature forecasts within wintertime cold air masses over, and downstream from, snow cover across the U.S. Great Plains. Likewise, this study illustrates the importance of the various components of the heat balance between the lower atmosphere and snow cover as based on the physical characteristics of the snowpack, which could prove beneficial in considerations of snow cover in weather and climate models.

Corresponding author address: Dr. Andrew W. Ellis, Department of Geography, Arizona State University, Box 870104, Tempe, AZ 85287-0104.

andrew.w.ellis.@asu.edu

Abstract

The presence of snow cover has been shown to modify atmospheric conditions through much of the earth’s troposphere due to its radiative effects. Snow cover has garnered much attention in recent decades as a result of concerns associated with potential changes in the global environment that may be intensified by the presence or absence of a snow cover. As a result, a greater emphasis has been placed on the representation of snow cover in weather and climate prediction models. This study investigates the effects of snow albedo and snow depth on the modification of surface air temperatures within cold air masses moving across the U.S. Great Plains in winter.

Through the adaptation of a one-dimensional snowpack model, the thermal characteristics of the core of a cold air mass were derived from the equation governing the heat balance between the surface and the lower atmosphere. The methodology was based on the premise that the core of a cold air mass may be considered homogeneous and not subject to advection of air from outside, thereby isolating the exchange of energy between the surface and the atmosphere as the control on lower-tropospheric temperatures. The adapted model included the synergism of the air mass–snow cover relationship through time, incorporating the natural feedback process.

Simulation of surface air temperatures within four cold air masses over snow cover of different albedo values and depths led to several conclusions. In testing the effects of snow albedo, results indicate 1) mean daytime air temperatures 3°–6°C higher and maximum daytime air temperatures 7°–12°C higher over snow with an albedo equal to 0.50 compared to 0.90, as a consequence of differences in sensible heat flux, and ultimately, absorbed solar radiation, and 2) little thermal inertia and therefore little difference in subsequent nighttime airmass temperatures over snow with an albedo of 0.50 compared to 0.90. In testing the effects of snow depth, results indicate 1) little difference in daytime air temperatures associated with a snow depth of 2.5 cm compared to 15.0 or 30.0 cm, 2) an increase in mean nighttime temperatures of 0.2°–0.7°C over a snow depth of 2.5 cm compared to either of the larger depths, and 3) a masking of the underlying bare soil surfaces by the snow depths of 15.0 and 30.0 cm and virtually no difference in airmass temperatures over the two snow depths.

The potential utility of the results of this study lies in their application as additional guidance for temperature forecasts within wintertime cold air masses over, and downstream from, snow cover across the U.S. Great Plains. Likewise, this study illustrates the importance of the various components of the heat balance between the lower atmosphere and snow cover as based on the physical characteristics of the snowpack, which could prove beneficial in considerations of snow cover in weather and climate models.

Corresponding author address: Dr. Andrew W. Ellis, Department of Geography, Arizona State University, Box 870104, Tempe, AZ 85287-0104.

andrew.w.ellis.@asu.edu

Introduction

Large changes in snow cover extent can result in the modification of atmospheric conditions through the earth’s lower troposphere due to the radiative effects of snow. In reflecting large amounts of incident shortwave radiation and absorbing heat through melting, it has been indicated that snow cover can lower surface air temperatures over timescales of days to months (e.g., Namias 1962; Dewey 1977). Snow cover has also been shown to affect the circulation of the atmosphere on various spatial scales through its influence on diabatic heating, which is determined by the extent, depth, and location of the snow cover (Robinson et al. 1995a). In recent decades scientists have produced evidence of links between snow cover and anomalous weather patterns, such as changes in precipitation patterns (Dey and Kumar 1983; Dickson 1984), temperature fields (Heim and Dewey 1984; Walsh et al. 1985; Namias 1985; Cerveny and Balling 1992; Leathers and Robinson 1993), cyclone frequencies and intensities (Ross and Walsh 1986; Dewey 1987), large-scale circulation patterns and 1000–500-mb atmospheric thickness (Lamb 1950; Namias 1978; Walsh et al. 1982; Barnett et al. 1989; Gutzler and Rosen 1992), and outbreaks of severe weather over areas remote from the snow-covered surface (Dewey 1987). At its peak in winter, snow covers approximately 46% of the land surface in the Northern Hemisphere (Robinson et al. 1995b). Changes in the snow cover climatology may act to intensify potential changes in the global environment through radiative and latent heat–induced feedback mechanisms, producing greater interest in snow cover–atmosphere interactions.

Several studies have investigated the importance of the presence of snow cover on local and regional atmospheric temperatures. Dewey (1977) compared forecasted maximum and minimum temperatures to measured values following a late season snowstorm over the upper Great Plains. Forecasted temperatures were 3°–15°C higher than measured with the forecast errors located over areas of anomalously deep snow cover. Similarly, Leathers et al. (1995) investigated temperature depressions associated with snow cover across the northeast United States. Depressions of average maximum and minimum temperatures of approximately 6°C and 5°C, respectively, were found for days characterized by snow cover of 2.5-cm depth or greater. Ellis and Leathers (1998) simulated cold airmass temperatures across the U.S. Great Plains under bare ground and snow cover conditions. Mean daytime lower-atmospheric airmass temperatures were found to be 6°–10°C higher over bare ground compared to snow cover, and maximum daytime temperatures were found to be 10°–15°C higher over bare ground. Mean nighttime temperatures were found to be only 1°–2°C higher over bare ground compared to snow cover.

Little research has addressed the roles of snow depth and snow albedo in modifying lower-atmospheric temperatures. Baker et al. (1992) compared maximum and minimum temperatures on days stratified by snow cover of at least 10-cm depth and snow cover of less than 10-cm depth to values on days without snow cover at St. Paul, Minnesota. Average values were 8.4°C lower for days with at least 10-cm snow depth than for days without snow cover. Average temperatures were 6.4°C lower for days with less than 10-cm snow depth than for days without snow cover. In addition, Baker et al. (1992) found surface temperatures to be 15°C and 10°C lower for snow covers of at least 10 cm and less than 10 cm, respectively, in comparison to surface temperatures for days with no snow cover. This study investigates the roles of snow depth and snow albedo in the modification of lower-tropospheric temperatures within four cold air masses moving across the U.S. Great Plains.

A one-dimensional snowpack model was altered in a manner such that the thermal characteristics of the core of a cold air mass could be derived solely from the equation governing the heat balance between the surface and the lower atmosphere. The methodology was based on the premise that the core of a cold air mass may be considered homogeneous and not subject to advection of air from outside the system. The exchange of energy between the atmosphere and the surface was effectively isolated as the control on lower-atmospheric temperature as the airmass core moved across the Great Plains. The adapted model included the synergism of the air mass–snow cover relationship through time, incorporating the thermal feedback process. The Great Plains study region was defined as the area within the United States west of 92°W longitude, east of the eastern slopes of the Rocky Mountains, north of 30°N latitude, and south of 49°N latitude (Fig. 1).

Snowpack model

A one-dimensional mass and energy balance model for a snowpack (SNTHERM; Jordan 1991) was adapted for use in this study. In predicting temperature profiles within snow and soil, the model addresses the full range of meteorological conditions common in winter, including freeze–thaw cycles and a full range of precipitation types. The transport of liquid water and water vapor is included in the model for use in the heat balance equations. The governing equations for heat and mass balance are applied to horizontally infinite control volumes, or one-dimensional layers of snow and soil, in order to obtain a numerical solution of the temperature profile of the snowpack. The boundary conditions for the model are determined by the meteorological conditions at the air–surface interface. Surface energy fluxes are calculated from measured meteorological data, which includes air temperature, relative humidity, wind speed, precipitation, and measured or calculated solar and incoming infrared radiation. The model is capable of calculating radiation values through routines that consider the solar altitude and azimuth, cloud conditions, surface albedo, and inclination of the surface (Jordan 1991). Profiles of temperature and water content for the various control volumes initialize the model, and the physical strata can be supplied by the user or extracted from internal databases for snow, sand, and clay.

The surface energy flux (I) within the model contains the turbulent fluxes for sensible and latent heat, shortwave and longwave radiation, and convected heat from precipitation (Jordan 1991), and can be written as
IRsαsRlRlH
where Rs↓ is the downward flux of energy from solar radiation, αs is the surface albedo (0.80 for snow, 0.35 for clay soil), Rl↓ and Rl↑ are the downward and upward components of longwave radiation, and H, LE, and CV are the turbulent fluxes of sensible, latent, and convective heat from falling precipitation. Model fluxes are defined as positive in the downward direction. A three-layer insolation model designed by Shapiro (1987) is used to estimate solar radiation (in lieu of measurements) and the downward flux of longwave radiation is calculated using the method of Idso (1981). The upward flux of longwave radiation can be written as
RlsσT40sRn
where εs is the emissivity of the surface (0.97 for snow, 0.90 for clay soil), σ is the Stefan–Boltzmann constant (5.669 × 10−8 W m−2 K−4), T0 is the surface temperature, and Rn↓ is the downward flux of shortwave radiation. The sensible heat flux is expressed as
HρaCpCHwTaT0
where ρa is air density, Cp is the specific heat of air at constant pressure, CH is the bulk transfer coefficient for sensible heat, w is wind speed at a height of 1.5 m, and Ta is air temperature at 1.5 m. The latent heat flux is expressed as
LυiCEwρυ,aρ0υ,sat
where Lυi is the latent heat of sublimation (2.838 × 106 J kg−1), CE is the bulk transfer coefficient for latent heat, ρυ,a is the vapor density in air, and ρ0υ,sat is the intrinsic water vapor density at saturation at the surface. The bulk transfer coefficients are calculated using the roughness length and are considered to be equal.

Methodology

Historical daily weather maps were examined in order to identify cases in which a wintertime cold air mass moved over the Great Plains as characterized by a well-defined surface synoptic weather pattern. Four such scenarios were chosen for study and they provided a range in the timing within the winter season of early January to early March (Table 1). For each cold airmass scenario, the daily snow depth across the study region was determined using data from 286 stations that formed a subset of the Historical Daily Climate Dataset (HDCD;Robinson 1993). The HDCD dataset is quality controlled based on various parameters that were designed to identify questionable data entries. Questionable data were eliminated from this study, as it is believed that the density and homogeneity of the station distribution (Fig. 2a) allowed for minor data elimination.

In order to characterize the atmosphere during the time period inclusive of each cold airmass scenario, hourly surface meteorological data were obtained for 57 stations distributed across the study region (Fig. 2b). Data were extracted from the EarthInfo, Inc., National Climatic Data Center Surface Airways database, which contains hourly or 3-hourly measurements of 15 meteorological variables. Data used within this study included air temperature, dewpoint temperature, sea level pressure, relative humidity, cloud-base height, fraction of sky covered by clouds, and wind speed. Each hour for which data were missing was assigned a value based on linear interpolation, for a period of up to three consecutive hours. Stations missing data for a period of greater than 3 h within the timing of a cold airmass scenario were eliminated from the study. Extended periods of cloud cover for which the less-frequently reported variable of cloud-base height was missing were filled using a convective cloud-base scheme (Ahrens 1994). Hourly meteorological data and daily snow depth data were interpolated to a 1° latitude by 1° longitude grid across the study region. Several additional parameters used to describe the physical characteristics of snow and soil were taken from previous work involving validation of the SNTHERM snowpack model (Jordan 1991).

Within the model the soil type was defined as clay, and the control volumes, or nodes, varied in thickness from 5 cm near the surface to 20 cm at a depth of over 1 m. The clay soil type was employed because it is believed to be a better representative of the finer soil particles of the Great Plains than the sand soil type. Sensitivity analyses showed that soil type had only a trivial effect on overlying snowpack and air temperatures. A snowpack was initially represented by at least two nodes totaling a thickness of 3 cm. The remaining depth of a snowpack was divided into a number of nodes that increased incrementally until the thickness of each node fell within the range of 0.8–5.0 cm. The initial temperature profiles for soil and snow were based on the hourly air temperature. A snowpack was initialized at each grid cell across the study region as isothermal and equal to the hourly surface air temperature. The temperature of the top soil node, with or without snow cover, was set equal to the hourly surface air temperature, and the temperature of the subsequent nodes below the surface were set equal to the temperature of the node above plus 0.5°C. For each hour within a cold airmass scenario and at each grid cell within the study region (every 1° × 1° box), the SNTHERM model processed the measured hourly meteorological data through three repetitive iterations. This was the minimum time length of model spinup that was found to bring the initial snowpack/soil temperature profile to equilibrium.

The core of a cold air mass was defined as a 5° latitude by 5° longitude box, composed of 25 1° latitude by 1° longitude grid cells (Fig. 3), surrounding the center of the anticyclone that is typically associated with a pure cold air mass. The small size definition of an airmass core reduced the likelihood of interference by advective processes in the attempt to isolate the energy flux across the air–surface interface as the controller of thermal variations within the airmass core. The airmass core was “moved” across the Great Plains as one coherent unit, with the hourly position determined by a linear change in latitude and longitude. The hourly, linear change was calculated using the timing and location of the anticyclone upon its entry and exit into and out of the study region. As cold air masses typically move south and east across the Great Plains in winter, the grid cells within the airmass core were indexed from the lower right (or southeast) to the upper left (or northwest).

Excluding the turbulent flux of convective heat from precipitation within a pure cold air mass (we assume no precipitation in the high pressure core), the model equation for the energy flux across the air/surface interface can be written as
Iair/surfRlnetH
where Rlnet is the net longwave radiation. The model energy flux equation used here is different from (1) in that the downward flux of solar radiation is omitted as it has no direct effect on air temperature. The effect of absorbed solar radiation on the surface temperature is reflected in air temperature through the turbulent heat fluxes and outgoing longwave radiation. The model equation for sensible heat flux can be written as
HCskρaCpCHSwTaT0
where Csk is a windless exchange coefficient for heat and S is a stability function. The model equation for latent heat flux can be written as
i1520-0450-38-6-696-e7
where Ck is a windless exchange coefficient for water vapor, Ce is a bulk transfer coefficient for water vapor at neutral stability, Rw is the gas constant for water vapor, ea is the vapor pressure of air, rhfo is the fractional humidity of the surface relative to steady state, and es is the vapor pressure at the surface.
Upon substituting the expanded terms for the fluxes of sensible and latent heat [(6) and (7)] into the energy flux equation (5), a quadratic function for air temperature can be derived in the form ax2 + bx + c = 0, where
i1520-0450-38-6-696-e8
and x = Ta, the air temperature. The quadratic formula can then be used to solve for air temperature.

With the stipulations of airmass core homogeneity and lack of advection, it was assumed that over short time intervals the energy flux between the lowest levels of a cold airmass core and the underlying surface would remain constant if not for variations in the surface conditions and insolation. After the initial conditions were applied to the first hour of the simulation period for a particular grid cell within the cold airmass core, hourly air temperatures were derived solely from the surface conditions and remaining meteorological conditions. For each hour (location), the simulated surface air temperature and measured meteorological conditions from the previous hour were combined with the surface conditions for the current hour in the original SNTHERM snowpack model. The results were an updated energy flux and an adjustment of the surface conditions, which were then combined with the current meteorological conditions in the revised model containing the quadratic function for air temperature. The energy flux of the previous time step would be duplicated if not for different surface conditions and insolation values, both of which change with time of day and location. The simulated airmass core temperatures subsequently altered the snowpack/soil temperature profile, initiating a feedback process between the snowpack and the overlying atmosphere.

For each of the four cold air masses studied, airmass core temperatures were simulated over the measured snow cover depth/extent for which the snow albedo value of 0.80 was used. The effect of snow albedo on the thermal characteristics of each cold air mass was tested using the two additional values of 0.90 and 0.50 to characterize the albedo of the measured snow cover. It was intended for the three albedo values to represent a pristine snow cover with no protrusions (0.90), a relatively typical snow cover (0.80), and a rather impure snow cover with many protrusions (0.50). Baker et al. (1991) found that an albedo of 0.70 or greater indicated a snow depth great enough to effectively mask the underlying surface. In order to test the sensitivity of the cold air mass to the depth of the underlying snow cover, airmass core temperature simulations were performed over uniform snowpacks of depths equal to 2.5 cm, 15 cm, and 30 cm covering the entire study region. In testing the sensitivity of the cold air mass to snowpack albedo and depth, the effect of each variable was artificially isolated. Relatively extreme snow albedo values were employed and the albedo values did not change through time and space. Snow depth was varied over a flat surface that offered no protrusions through the snow surface, regardless of the snow depth. Any temperature modification would be the direct result of the specific snow depth or snow albedo employed. Model evaluation statistics were compiled in order to measure the agreement between simulated and measured airmass core temperatures. Time series of temperature and the various energy fluxes were constructed in order to reveal the diurnal trends and latitudinal influences associated with the relationships between cold airmass temperatures and the depth and albedo of the underlying snow cover. Finally, spatial distributions of temperature within each cold air mass were analyzed in order to reveal the nature of the feedback process between the air mass and the surface.

Results and discussion

Simulations over measured snow cover

Each of the four cold air masses chosen for study was characterized by a leading cold front, low air and dewpoint temperatures, and high sea level pressures associated with an anticyclone that moved in a southerly to southeasterly direction across the Great Plains. In each case, a stationary front developed over the western extent of the study region, along the eastern slopes of the Rocky Mountains. The length of the time period for which the cold airmass core of each of the four scenarios was positioned within the study region ranged from 36 to 49 h. The airmass cores associated with scenarios one through three remained over snow cover for the duration of each simulation period (Figs. 3a–c). The airmass core associated with scenario four was positioned over bare ground for approximately the final 12 h of the simulation period (Fig. 3d).

Average simulated airmass core temperatures, calculated from all of the values within the airmass core at each time step, compare favorably to measured values through the time period of each simulation (Figs. 4a–d). In each case the diurnal trend in temperature was simulated well, as was the general increase in temperature with time, which corresponded with the movement of the air masses into lower latitudes. Model evaluation statistics (Table 1) describe the overall performance of the simulation procedure. Mean airmass core temperatures taken from the full simulation period of each scenario reveal that differences between measured and simulated values are generally less than 0.75°C, while standard deviations differ by less than 1.10°C. High correlation coefficient and index of agreement values indicate a very good interrelation between the simulated and measured temperatures in terms of variability and magnitude (Willmott 1984). Low root-mean-square error and mean absolute error values further indicate a good overall agreement, with a large percentage of the error being unsystematic or random (Table 1).

Simulations over measured snow cover of different albedoes

Decreasing the uniform albedo of the measured snow cover across the study region from 0.90 to 0.50 produced considerable warming of the daytime temperatures associated with each cold airmass core. Differences in daytime mean airmass core temperatures range from approximately 1°–2°C early and late in the daytime period to 7°–12°C during peak insolation hours (Figs. 5a–d). The most noticeable warming occurred within the cold airmass core of scenario four (Fig. 5d), which is most likely attributable to its later timing within the winter season (29 Feb–2 Mar), and associated greater insolation. Differences in the mean daytime airmass core temperatures produced by the snow albedo values of 0.90 and 0.50 through the full simulation period of each scenario range from slightly greater than 3°C to nearly 6°C (Table 2). Daytime differences between the temperatures produced by the snow albedo values of 0.50 and 0.80 (control albedo) ranged from nearly 2.5°C to greater than 4°C (Table 2). Temperature differences from throughout the nighttime hours of each scenario are virtually nonexistent (Figs. 5a–d), with any difference attributable to very small temperature differences in the hours immediately following sunset. It is apparent that very little thermal inertia from daytime hours exists in the cold air mass–snow cover relationship. The cold, dry air masses over the Great Plains in winter offer very little resistance to nighttime heat and radiation loss from the snowpacks.

Boundary layer energy flux values help to explain the differences in the airmass core temperatures produced by the simulations involving the different snow albedo conditions. The mean airmass core fluxes of sensible heat were directed upward (−) during the midday hours of scenarios two and three for snow albedo values of 0.50 and 0.80 (Figs. 6b and 6c). Upward-directed fluxes of sensible heat contribute to warmer airmass core temperatures, and in this case a larger contribution was made over snow with an albedo value of 0.50 as represented by sensible heat fluxes of larger absolute magnitudes. Within scenarios two and three, the daytime sensible heat fluxes associated with a snow albedo value of 0.90 were generally directed downward (+), except during a few midday hours of scenario two (Figs. 6b and 6c), which indicates a contribution to cooling the overlying atmosphere. During the midday hours of scenarios one and four, upward-directed (−) sensible heat flux values were only produced by a snow albedo of 0.50 (Figs. 6a and 6d). The daytime fluxes of sensible heat associated with the higher albedo values were consistently directed downward (+), contributing to cooler atmospheric temperatures, and to a greater degree with an albedo equal to 0.90. Differences in the magnitudes and the directions of the daytime sensible heat fluxes associated with scenarios one and four compared to scenarios two and three are reflected in the mean values through each full simulation period (Table 2). The cold airmass cores of scenarios two and three are simply colder than those of scenarios one and four (Table 2), and the result is a marked difference in sensible heat flux magnitude and, in some instances, the direction of the flux. The nighttime fluxes of sensible heat associated with each albedo value and within each scenario were virtually identical and consistently directed downward (+; Figs. 6a–d). Only very small differences can be found immediately after sunset, where an albedo value of 0.50 produces the greatest amount of downward-directed sensible heat flux. The greater fluxes associated with the lower albedo values contribute to a greater cooling of the overlying atmosphere and a rapid drop in the airmass core temperatures (Figs. 5a–d).

The means of the latent heat flux values within the airmass cores through the time periods of simulation indicate fluxes that were almost always directed upward (−) for each cold airmass scenario, regardless of snow albedo (Figs. 7a–d). These are indicative of vapor fluxes that were directed upward toward the overlying dry air masses. Only occasionally during nighttime hours was the direction of the flux reversed, and then only for a brief time period and of a small magnitude. During daytime hours, the absolute magnitudes of the latent heat fluxes were greatest over snow with an albedo of 0.50 and smallest over snow with an albedo of 0.90 (Table 2). This is likely attributable to the higher snow surface temperatures associated with the lower snow albedo, and a consequently greater availability of moisture at the surface through melting or sublimation processes. In such a case, the greater amount of evaporation from the warmer snow surface (α = 0.50) could eventually contribute to warmer lower atmospheric temperatures if condensation of the atmospheric water vapor were to occur. This would be of additional importance during cooler nighttime hours when increased atmospheric moisture might have a noticeable effect on net longwave radiation. The differences in the daytime latent heat fluxes produced by the different snow albedo conditions were greater at lower latitudes (Figs. 7a–c) and later in the winter season (Fig. 7d). During nighttime hours, the absolute magnitudes of the latent heat fluxes were much smaller, and the differences between the fluxes produced by the various snow albedo values were small. Lower nighttime temperatures within each airmass core, combined with the lack of insolation, led to a frozen snowpack surface.

The mean values of net longwave radiation within the airmass cores through the time periods of simulation indicate net fluxes that were always directed upward (−) within each of the four cold airmass cores, regardless of snow albedo (Figs. 8a–d). Through the time periods of simulation, outgoing longwave radiation exceeds incoming longwave radiation by as little as 5 W m−2 and as much as 90 W m−2. During daytime hours, the net fluxes of longwave radiation within the airmass cores were generally of the greatest absolute magnitudes over snow with an albedo of 0.50, and of the smallest absolute magnitudes over snow with an albedo equal to 0.90 (Figs. 8a–d; Table 2). This simply reflects the warmest snow surface temperatures that are associated with the lowest snow albedo. During nighttime hours, the absolute magnitudes of the fluxes were nearly identical, again reflecting the virtual absence of thermal inertia from daytime hours. Overall, differences in the values produced by the different surface conditions were relatively small, and therefore net longwave radiation appears to be less significant in the modification of cold airmass core temperatures than the fluxes of sensible and, potentially, latent heat.

Daytime differences in the various energy fluxes across the snow–air interface that were produced by the different snow albedo conditions were largely the result of snowpack temperature differences produced by variations in the amount of solar radiation absorbed (Table 2). The effect on the daytime processes was prominent but, as evidenced, that effect was quickly dampened after sunset. The flux of energy across the snow–soil interface also modifies snowpack temperatures and, subsequently, the energy fluxes across the snow–air interface. In nearly each case, the mean daytime flux of energy across the snow–soil interface was directed away from the snowpack (−; toward the underlying soil) and of the greatest absolute magnitude for a snowpack with an albedo value of 0.50 (Table 2). This indicates that the greatest, yet very small, amount of energy was transferred to the underlying soil from a snowpack with an albedo of 0.50 during the day, and the least was transferred from a snowpack with an albedo of 0.90. This is simply a reflection of the warmer daytime temperatures associated with the lower snow albedo. The very low values of the energy fluxes across the snow–soil interface likely do not contribute greatly to snowpack temperature variations and, in turn, lower atmospheric temperatures.

The spatial distributions of the differences between the mean daytime airmass core temperatures produced by the more extreme snow albedo values of 0.50 and 0.90 show the feedbacks between the air masses and the underlying surfaces. Calculated at each individual grid cell using airmass core temperatures for each hour of daylight within the full period of each simulation (during which time the air mass has moved a considerable distance), differences are presented as contour lines across the generic 5° × 5° airmass core (Figs. 9a–d). Mean airmass core temperature differences generally increase from southeast to northwest along the axes of the airmass movements. In the case of the snow albedo equal to 0.50, the northwestern quadrants of the airmass cores were continually modified by underlying surfaces that had been previously influenced by the southeastern quadrants of the anomalously warm airmass cores. The results were enhanced modifications of the upstream airmass core temperatures, due to the synergism in which the airmass core temperatures were modified by the underlying surface and vice versa.

Simulations over uniform snow cover of different depths

Uniformly varying the depth of a snow cover across the entire study region produced very small modifications to the cold airmass core temperatures, and primarily only during nighttime hours. Due to the small differences, time series of simulation output associated only with cold airmass scenario one are presented. Consistently, very little difference (<0.1°C) exists between daytime or nighttime airmass core temperatures produced over a snow depth of 15.0 cm compared to a snow depth of 30.0 cm. Daytime mean airmass core temperatures over a snow depth of either 15.0 or 30.0 cm are approximately 0.1°C warmer than temperatures over a snow depth of 2.5 cm (Fig. 10a). During nighttime hours, mean airmass core temperatures are generally 0.5°C–1.0°C warmer over a snow depth of 2.5 cm than over a snow depth of either 15.0 or 30.0 cm (Fig. 10a). Differences between the mean nighttime airmass core temperatures produced by the smaller snow depth and the larger snow depths through the full simulation period range from 0.2°C to 0.7°C (Table 3). Again, it is important to note that each snowpack is situated on a flat surface that offers no protrusions through the snowpack. The surface of each snowpack is alike, regardless of the snow depth. Also, it should be noted that the small temperature differences (Table 3) are of a similar magnitude of those differences between simulated and measured temperatures (Table 1).

Mean values for the energy fluxes across the snow–air interface and the snow–soil interface help to explain the modest differences in the airmass core temperatures produced by the different snow depths. The sensible heat fluxes over the thinner snowpacks were generally of smaller magnitudes than were those over the larger snow depths (Fig. 10b; Table 3), contributing to slightly lower airmass temperatures.

The means of the latent heat flux values within the airmass cores throughout the time periods of simulation (not shown) indicate fluxes that were almost always directed upward (−) for each cold airmass scenario, regardless of snow depth. Only occasionally during nighttime hours was the direction of the flux reversed. The differences between the latent heat fluxes produced by the different snow depths are very small (Fig. 10c; Table 3) and would only minimally affect surface air temperatures if the atmospheric water vapor were to condense.

The mean values of net longwave radiation within the airmass cores through the time periods of simulation (not shown) again indicate net fluxes that were consistently directed upward (−) within each of the four cold airmass cores, in this case, regardless of snow depth. Differences between the mean values produced by each snow depth are small, indicating little contribution to the small temperature differences (Fig. 10d; Table 3).

With surfaces that are physically consistent between each snow depth, net solar radiation is not a cause of snowpack temperature variations by snow depth. Instead, the flux of energy across the snow–soil interface is of greater significance. The energy flux is directed from the snowpack toward the frozen soil (−) during daytime hours for snowpacks of 2.5- and 15.0-cm depth (Figs. 11a–d), and is greater with a snow depth of 2.5 cm. This indicates a greater heat flux from the thinner snowpack into the frozen soil during daytime hours and a contribution to cooler snowpack temperatures. During nighttime hours, the energy flux is directed toward the snowpack from the underlying soil (+) for each snow depth and is greatest for the smallest snow depth (Figs. 11a–d; Table 3). This indicates a greater contribution to warmer nighttime snowpack temperatures within the thinner snowpack.

Despite relatively large differences in the magnitude and direction of the heat fluxes across the snow–soil interfaces with snow depths of 15.0 and 30.0 cm, the energy fluxes across the snow–air interfaces and the temperatures of the overlying cold air masses show virtually no indication of the difference. The thicker snow depths seem to effectively mask the underlying soil. This agrees with the findings of Baker et al. (1991) who concluded that a snow depth of 5.0 cm or greater is necessary to mask an underlying bare soil surface. The flux of sensible heat appears to be the primary process that contributes to the slightly warmer nighttime temperatures over the thinner snowpack, and this appears to be mainly the result of a greater flux of heat from the soil into the thinner snowpack during nighttime hours.

Conclusions

This study investigated the effects of snow albedo and snow depth on surface air temperatures within four wintertime cold air masses moving across the U.S. Great Plains. A one-dimensional snowpack model was adapted in a manner such that the thermal characteristics of the core of a cold air mass could be derived from the equation governing the heat balance between the surface and the lower atmosphere. This allowed for the isolation of the energy exchange between the atmosphere and the surface as the control on lower atmospheric temperature. The procedure produced good agreement between simulated and measured airmass core temperatures in terms of both variability and magnitude.

In decreasing the uniform albedo of the measured depth of snow across the study region from 0.90 to 0.50, results indicate 1) increases in mean daytime temperatures of 3°–6°C and increases in maximum daytime temperatures by 7°–12°C as a consequence of differences in sensible and, potentially, latent heat fluxes, which were ultimately consequences of differences in absorbed solar radiation, and 2) little or no increase (less than 0.1°C) of mean nighttime temperatures and a very rapid nighttime cooling of the snowpacks.

In uniformly varying the depth of snow cover across the entire study region, results show 1) virtually no difference in airmass core temperatures over a snow depth of 15.0 cm compared to 30.0 cm; 2) very small increases in daytime temperatures by less than 0.1°C over either of the larger snow depths compared to a snow depth of 2.5 cm; 3) increases in mean nighttime temperatures by 0.2°–0.7°C over a snow depth of 2.5 cm compared to either of the larger depths, primarily as a consequence of sensible heat flux differences; 4) an evident contribution of the heat flux across the snow–soil interface to the increases in nighttime temperatures over the thinner snowpacks; and 5) relatively large differences in the heat fluxes across the snow–soil interfaces of the two larger depths that did not affect the temperature of the overlying air masses. This agrees with the findings of Baker et al. (1991), who concluded that snow depths of 5.0 cm or greater effectively mask an underlying bare soil surface.

Accurate parameterization of snow cover within forecast models could result in a greater accuracy in the prediction of cold airmass temperatures and assistance in forecasting such phenomena as frost/freeze events, snow-breeze occurrence, intensification and trajectory of midlatitude cyclones, lake-effect snowfall, and potentially early spring severe weather outbreaks. The methodology that is employed in this research is distinguished from previous research by the emphasis placed on the detailed description of the underlying surface and the feedback process between the surface and the air mass.

Recently completed research simulated the effects of the presence of snow cover, or snow cover extent, on atmospheric temperatures across the U.S. Great Plains (Ellis and Leathers 1998). Leathers et al. (1995) noted that the influence of snow cover on the lower atmosphere is likely to be significantly different over regions that are more physically diverse than the Great Plains, in terms of terrain, vegetation, and location relative to large water bodies. In addition to supplemental research involving the interactions between snow cover and a larger number of cold air masses and in regions that are more diverse, investigation into several additional topics could prove to be useful. These topics include the effects of variations in 1) the intrannual timing of the cold air mass, 2) the trajectory and/or speed of movement of the cold air mass, and 3) the zonality in the atmospheric circulation pattern (the degree of negative vorticity advection). Researching the effects of snow cover on cold air masses developing over their source regions could potentially provide assistance in longer-lead forecasts of cold airmass temperatures. A more advanced method for quantifying the interaction between snow cover and cold air masses that includes temperature changes through advection is currently being researched.

Acknowledgments

The authors would like to thank Brian Hanson, Laurence Kalkstein, and David Robinson for their thoughtful comments regarding the methodology of this research and for their various contributions to the datasets used. Additional appreciation is extended to Rachel Jordan, Clint Rowe, and Andrew Grundstein for their help with various facets of the snowpack model. This work was partially funded by the National Science Foundation under Grant SBR-9320786.

REFERENCES

  • Ahrens, C. D., 1994: Meteorology Today. West Publishing, 591 pp.

  • Baker, D. G., R. H. Skaggs, and D. L. Ruschy, 1991: Snow depth required to mask the underlying surface. J. Appl. Meteor.,30, 387–392.

  • ——, D. L. Ruschy, R. H. Skaggs, and D. B. Wall, 1992: Air temperature and radiation depression associated with a snow cover. J. Appl. Meteor.,31, 247–254.

  • Barnett, T. P., L. Dumenil, U. Schlese, E. Roeckner, and M. Latif, 1989: The effect of Eurasian snow cover on regional and global climate variations. J. Atmos. Sci.,46, 661–685.

  • Cerveny, R. S., and R. C. Balling, 1992: The impact of snow cover on diurnal temperature readings. Geophys. Res. Lett.,19, 797–800.

  • Dewey, K. F., 1977: Daily maximum and minimum temperature forecasts and the influence of snow cover. Mon. Wea. Rev.,105, 1594–1597.

  • ——, 1987: Snow cover–atmosphere interactions. Large-Scale Effects of Seasonal Snow Cover: Proceedings of the Vancouver Symposium, Int. Assoc. Hydrol. Sci. Publ. 166, 27–42.

  • Dey, B., and B. Kumar, 1983: Himalayan winter snow cover area and summer monsoon rainfall over India. J. Geophys. Res.,88, 5471–5474.

  • Dickson, R. R., 1984: Eurasian snow cover versus Indian monsoon rainfall—An extension of the Hahn–Shukla results. J. Climate Appl. Meteor.,23, 171–173.

  • Ellis, A. W., and D. J. Leathers, 1998: A quantitative approach to evaluating the effects of snow cover on cold air mass temperatures across the U.S. Great Plains. Wea. Forecasting,13, 688–701.

  • Gutzler, D. S., and R. D. Rosen, 1992: Interannual variability of wintertime snow cover across the northern hemisphere. J. Climate,5, 1441–1447.

  • Heim, R., Jr., and K. F. Dewey, 1984: Circulation patterns and temperature fields associated with extensive snow cover on the North American continent. Phys. Geogr.,4, 66–85.

  • Idso, S. B., 1981: A set of equations for full spectrum and 8–14 μm and 10.5–12.5 μm thermal radiation from cloudless skies. Water Resour. Res.,17, 295–304.

  • Jordan, R., 1991: A one dimensional temperature model for a snow cover: Technical documentation for SNTHERM.89. Special Rep. 91-16, U.S. Army Corps of Engineers, Cold Regions Research and Engineering Laboratory, Hanover, NH. [Available from U.S. Army Corps of Engineers, Cold Regions Research and Engineering Laboratory, Hanover, NH 03755-1290.].

  • Lamb, H. H., 1950: Types and spells of weather around the year in the British Isles: Annual trends, seasonal structure of the year, singularities. Quart. J. Roy. Meteor. Soc.,76, 393–438.

  • Leathers, D. J., and D. A. Robinson, 1993: The association between extremes in North American snow cover extent and United States temperatures. J. Climate,6, 1345–1355.

  • ——, A. W. Ellis, and D. A. Robinson, 1995: Characteristics of temperature depressions associated with snow cover across the northeast United States. J. Appl. Meteor.,34, 381–390.

  • Namias, J., 1962: Influences of abnormal surface heat sources and sinks on Atmospheric behavior. Proceedings of the International Symposium on Numerical Weather Prediction, 1960, Meteorological Society of Japan, 615–627.

  • ——, 1978: Multiple causes of the North American abnormal winter 1976–77. Mon. Wea. Rev.,106, 279–295.

  • ——, 1985: Some empirical evidence for the influence of snow cover on temperature and precipitation. Mon. Wea. Rev.,113, 1542–1553.

  • Robinson, D. A., 1993: Historical daily climatic data for the United States. Preprints, Eighth Conf. on Applied Climatology, Anaheim, CA, Amer. Meteor. Soc., 264–269.

  • ——, A. Frei, D. J. Leathers, and M. C. Serreze, 1995a: Northern Hemisphere snow cover during the transition seasons. Proc. 19th Annual Climate Diagnostics Workshop, College Park, MD, NOAA, 377–380.

  • ——, ——, and M. C. Serreze, 1995b: Recent variations and regional relationships in northern hemisphere snow cover. Ann. Glaciol.,21, 71–76.

  • Ross, B., and J. E. Walsh, 1986: Synoptic scale influence of snow cover and sea ice. Mon. Wea. Rev.,114, 1795–1810.

  • Shapiro, R., 1987: A simple model for the calculation of the flux of direct and diffuse solar radiation through the atmosphere. ST Systems Corporation Scientific Rep. 35, Lexington, MA, 49 pp. [Available from Rachel Jordan, Cold Regions Research Engineering Laboratory, Hanover, NH 03755-1290.].

  • Walsh, J. E., D. R. Tucek, and M. R. Peterson, 1982: Seasonal snow cover and short-term climatic fluctuations over the United States. Mon. Wea. Rev.,110, 1474–1485.

  • ——, W. H. Jasperson, and B. Ross, 1985: Influence of snow cover and soil moisture on monthly air temperature. Mon. Wea. Rev.,113, 756–768.

  • Willmott, C. J., 1984: On the evaluation of model performance in physical geography. Spatial Statistics and Models, G. L. Gaile and C. J. Willmott, Eds., D. Reidel, 443–460.

Fig. 1.
Fig. 1.

The study region encompassing the U.S. Great Plains.

Citation: Journal of Applied Meteorology 38, 6; 10.1175/1520-0450(1999)038<0696:AOCATM>2.0.CO;2

Fig. 2.
Fig. 2.

The station distributions for (a) snow depth and (b) meteorological data.

Citation: Journal of Applied Meteorology 38, 6; 10.1175/1520-0450(1999)038<0696:AOCATM>2.0.CO;2

Fig. 3.
Fig. 3.

The movement of airmass cores one through four (a)–(d), with the positions based on an interval of 12 h. The snow cover extent (≥2.5 cm) for the time period associated with each cold airmass core is also indicated with a heavy line.

Citation: Journal of Applied Meteorology 38, 6; 10.1175/1520-0450(1999)038<0696:AOCATM>2.0.CO;2

Fig. 4.
Fig. 4.

Mean simulated and measured average airmass core temperatures (°C) through the time periods of simulation for cold airmass scenarios one through four (a)–(d).

Citation: Journal of Applied Meteorology 38, 6; 10.1175/1520-0450(1999)038<0696:AOCATM>2.0.CO;2

Fig. 5.
Fig. 5.

Mean simulated airmass core temperatures (°C) produced by snow albedo conditions of 0.90, 0.80, and 0.50 through the time periods of simulation for cold airmass scenarios one through four (a)–(d).

Citation: Journal of Applied Meteorology 38, 6; 10.1175/1520-0450(1999)038<0696:AOCATM>2.0.CO;2

Fig. 6.
Fig. 6.

Mean simulated airmass core sensible heat flux values (W m−2) produced by snow albedo conditions of 0.90, 0.80, and 0.50 through the time periods of simulation for cold airmass scenarios one through four (a)–(d).

Citation: Journal of Applied Meteorology 38, 6; 10.1175/1520-0450(1999)038<0696:AOCATM>2.0.CO;2

Fig. 7.
Fig. 7.

Same as Fig. 6 but for latent heat flux.

Citation: Journal of Applied Meteorology 38, 6; 10.1175/1520-0450(1999)038<0696:AOCATM>2.0.CO;2

Fig. 8.
Fig. 8.

Same as Fig. 6 but for net longwave radiation.

Citation: Journal of Applied Meteorology 38, 6; 10.1175/1520-0450(1999)038<0696:AOCATM>2.0.CO;2

Fig. 9.
Fig. 9.

The spatial distribution of the difference between the mean daytime airmass core temperatures (°C) associated with snow albedo conditions of 0.50 and 0.90 throughout the full simulation periods for scenarios one through four (a)–(d). Differences are taken as temperatures associated with an albedo of 0.50 minus temperatures associated with an albedo of 0.90.

Citation: Journal of Applied Meteorology 38, 6; 10.1175/1520-0450(1999)038<0696:AOCATM>2.0.CO;2

Fig. 10.
Fig. 10.

(a) Mean simulated airmass core temperatures (°C), (b) sensible heat fluxes (W m−2), (c) latent heat fluxes (W m−2), and (d) net longwave radiation values (W m−2) produced by snow depth conditions of 2.5, 15.0, and 30.0 cm through the time period of simulation for cold airmass scenario one.

Citation: Journal of Applied Meteorology 38, 6; 10.1175/1520-0450(1999)038<0696:AOCATM>2.0.CO;2

Fig. 11.
Fig. 11.

Mean simulated ground heat flux values (W m−2) produced by snow depth conditions of 2.5, 15.0, and 30.0 cm through the time periods of simulation for cold airmass scenarios one through four (a)–(d).

Citation: Journal of Applied Meteorology 38, 6; 10.1175/1520-0450(1999)038<0696:AOCATM>2.0.CO;2

Table 1.

The model evaluation statistics that compare the simulated (s) and measured (m) mean airmass core temperatures through the time periods of cold airmass scenarios one through four. Included are the mean temperature (Ta; °C), the standard deviation (S; °C), the correlation coefficient (r), the index of agreement (D), the root-mean-square error (rmse), the mean absolute error (MAE), and the systematic (sy) and unsystematic (u) values of the rmse.

Table 1.
Table 2.

Daytime airmass core means under the snow albedo conditions of 0.90, 0.80, and 0.50 for the full simulation periods of cold airmass scenarios one through four. Included are surface air temperature (Ta), sensible heat flux (H), latent heat flux (LE), net longwave radiation (Rnlw), net solar radiation (Sn), and ground heat flux (G). Temperatures are in degrees Celsius and energy fluxes are in watts per meter squared.

Table 2.
Table 3.

Nighttime airmass core means under snow depth conditions of 2.5, 15.0, and 30.0 cm for the full simulations periods of scenarios one through four. Included are surface air temperature (Ta), sensible heat flux (H), latent heat flux (LE), net longwave radiation (Rnlw), and ground heat flux (G).

Table 3.
Save