Abstract

The relationship among the surface albedo, cloud properties, and radiative fluxes is investigated for the first time using a year-long cloud-resolving model (CRM) simulation with the prescribed evolving surface albedo. In comparison with the run using a fixed surface albedo, the CRM with the observed surface albedo represents the shortwave radiative budget closer to the observations in the winter. The greater surface albedo induces weaker instability in the low troposphere so that the amount of low clouds decreases during the winter. This reduces the shortwave and longwave cloud radiative forcing at the surface. The analysis of the CRM simulations with the evolving surface albedo reveals that there is a critical value (0.35) of the surface albedo. For albedos greater than the critical value, the upward shortwave flux at the top of the atmosphere (TOA) is positively proportional to the surface albedos when optically thin clouds exist, and is not much affected by reflection on the cloud top. If optically thick clouds occur and the surface albedo is greater than the critical value, the upward shortwave flux at the TOA is significantly influenced by the reflection of cloud top, but not much affected by the surface albedo. In addition, for albedos larger than the critical value, the downward shortwave flux at the surface is primarily influenced by the surface albedo and the reflection from the cloud base if optically thick clouds occur. However, the downward shortwave flux at the surface is not significantly affected by the surface albedo when optically thin clouds exist because the reflection on the cloud base is weak. When surface albedos are less than the critical value, those relationships among surface albedo, shortwave flux, and cloud properties are not obvious. The surface albedo effect on shortwave flux increases as solar zenith angle (SZA) decreases, but its dependence on the SZA is negligible when optically thick clouds exist.

1. Introduction

Surface albedo plays an essential role in determining the energy budget at the surface and the top of the atmosphere (TOA). Most albedo-related studies have focused on snow–albedo feedback and its impacts on climate sensitivity in general circulation model (GCM) simulations (Schneider and Dickinson 1974; Randall et al. 1994; Hall 2004; Winton 2006; Qu and Hall 2007). In addition, studies have been conducted over high-latitude regions such as Alaska, Greenland, and other Arctic and Antarctic regions in order to examine the effects of surface albedo and clouds on ultraviolet (UV) radiation through comparing snow-covered and snow-free areas and to quantify its temporal and spatial characteristics (Baker and Ruschy 1989; Stamnes et al. 1990; McKenzie et al. 1998; Kylling et al. 2000). For instance, Huber et al. (2004) quantified the effect of horizontal inhomogeneity of surface albedo on diffuse UV radiation at the High Alpine Research Station in Jungfraujoch, Switzerland, using a discrete ordinate radiative transfer model. There have also been some studies on desert albedo using satellite data. Tsvetsinskaya et al. (2002) reported that satellite data have convincingly shown the considerable spatial variation of desert albedo by analyzing the Moderate Resolution Imaging Spectroradiometer (MODIS) retrievals. Wang et al. (2005) also found that bare soil albedo is a function not only of soil color and moisture but also of solar zenith angle (SZA) using the MODIS bidirectional reflectance distribution function (BRDF) and albedo data over 30 desert regions.

Those previous studies have usually more focused on the effects on UV radiation rather than the relationship between surface albedo and clouds. Nichol et al. (2003), however, showed that a great amount of surface albedo can moderate the attenuation of UV radiation by cloud through the multiple scattering between the cloud base and the surface in the high latitudes. Shupe and Intrieri (2004) also examined the relationship between surface albedo and cloud radiative forcing (CF) over an Arctic region using the cloud and radiation dataset from the Surface Heat Budget of the Arctic (SHEBA) program. For middle latitude cases, some research groups have investigated various surface-albedo-related phenomena. Grant et al. (2000) examined the dependence of clear-sky albedo on the SZA by observing the daily variation of surface albedo at Uardry in southeastern Australia. Considering the impact of observed surface albedo over the Department of Energy Atmospheric Radiation Measurement Program (ARM) Southern Great Plains (SGP) site during two winter seasons, Dong (2005) improved a parameterization for low-level cloud properties. The parameterization was able to represent the effect of surface albedo on the stratus cloud microphysical and radiative properties. Duchon and Hamm (2006), analyzing observed daily broadband surface albedo over the ARM SGP for two years (1998 and 1999), reported that there is obvious horizontal inhomogeneity of surface albedo among the six observational ground stations: surface albedo over bare soil is significantly affected by precipitation but vegetated surfaces are not strongly affected. They also found that on an overcast day surface albedo tends to decrease. Yang et al. (2006) used direct and diffuse surface albedo produced from the ARM SGP and tropical western Pacific (TWP) sites during 1997–2004 to evaluate the parameterization of the dependence of surface albedo on SZA used by the National Centers for Environmental Prediction (NCEP) Global Forecast Systems (GFS) and those derived from satellite observations. Recently, Yang et al. (2008) showed the dependence of a snow-free surface albedo on SZA using the surface albedo data obtained during 1997–2005 from nine measurement stations whose surface types and locations are very different from each other.

During the past decade, cloud-resolving models (CRMs) have been widely used to simulate cloud systems under various large-scale conditions and over different climate regions. For example, Grabowski et al. (1996) conducted a weeklong CRM simulation adopting the evolving large-scale temperature and moisture advections during phase III of the Global Atmospheric Research Program (GARP) Atlantic Tropical Experiment (GATE) and showed three distinct simulated cloud regimes (i.e., nonsquall clouds, squall lines, and scattered convection) comparable to radar observations. Other monthlong CRM simulations have also conducted by Wu et al. (1998, 1999) during the Tropical Ocean Global Atmosphere Coupled Ocean–Atmosphere Response Experiment (TOGA COARE) between November 1992 and February 1993 and by Wu et al. (2007) during the ARM SGP intensive observation period (IOP) in 1997. They showed that the CRM-produced data have reasonable agreement with ground-measured and satellite-retrieved data such as temperature, moisture, surface heat fluxes, precipitation, and radiative fluxes at the surface and TOA. Recently, a yearlong CRM simulation was conducted over the ARM SGP for the year 2000 by Wu et al. (2008) to investigate the seasonal variation of radiative and cloud properties. The yearlong simulation provides a physically consistent long-term dataset for understanding the processes of convection, cloud, and radiation and the interaction among them. However, a large discrepancy of net shortwave (SW) flux between the CRM and observation at the surface is present in wintertime due to the use of fixed surface albedo.

The relationship among the surface albedo, radiative fluxes, and cloud properties cannot be investigated with the CRM using the fixed surface albedo. If the surface albedo in the CRM is varied diurnally and seasonally, the relationship would be revealed. In this study, a yearlong simulation is performed over the ARM SGP site for the year 2000 using the Iowa State University (ISU) CRM with the prescribed evolving surface albedo from the ARM observational estimates. The objectives are 1) to examine the effects of the evolving surface albedo on the CRM simulation and 2) to investigate the relationship among the surface albedo, cloud properties, and radiative fluxes. In the next section, the ISU CRM is briefly described and the prescription for the evolving surface albedo is explained. Annual and seasonal characteristics of simulated cloud properties and radiative fluxes are presented in section 3. In section 4 the seasonal variation of surface albedo and its effects on cloud properties and radiative fluxes are examined, and how the effect of surface albedo on radiative fluxes depends on SZA is investigated. A summary and discussion is given in section 5.

2. Cloud-resolving models and the prescribed evolving surface albedo

The ISU CRM is originally from a two-dimensional (2D) version of the Clark–Hall anelastic cloud model (e.g., Clark et al. 1996) with modified physical processes significant for the long-term simulations of cloud systems (e.g., Grabowski et al. 1996; Wu et al. 1998, 1999, 2007, 2008; Wu and Moncrieff 2001). The Kessler (1969) bulk warm rain parameterization and the Koenig and Murray (1976) bulk ice parameterization are adopted in the microphysical processes of the model. The ice parameterization produces two types of ice particles; type-A ice is slowly falling, low-density (unrimed or lightly rimed) particles and type-B ice is fast-falling, high-density particles such as graupel. Each type of ice is represented by two variables: mixing ratio and number concentration. For the radiative transfer calculation, the radiation scheme of the NCAR Community Climate Model (CCM) (Kiehl et al. 1996) is adopted with the use of liquid and type-A ice clouds, which have effective radii of 10 and 30 μm, respectively. The first-order eddy diffusion method of Smagorinsky (1963) is applied to parameterize the subgrid-scale mixing. Periodic lateral boundary conditions are adopted to facilitate a mathematically consistent CRM framework (Grabowski et al. 1996). The CRM uses the prescribed evolving surface sensible and latent heat fluxes from observations. To distribute the surface fluxes within the boundary layer, a nonlocal vertical diffusion scheme is used (Troen and Mahrt 1986; Holtslag and Moeng 1991; Hong and Pan 1996). The ISU CRM uses a 2D east–west 600-km horizontal and 40-km vertical domain. The horizontal grid size is 3 km and the vertical grid of 52 levels is 100 m at the surface, 550–850 m between 5 and 12 km, and 1500 m at the model top. The simulation time step is 15 s. The model setting of the yearlong (3 January–31 December 2000) CRM simulation is basically the same as the CRM simulation by Wu et al. (2008) except that the CRM in this study uses the prescribed evolving surface albedo for radiative calculation [hereafter, the CRM simulation by Wu et al. (2008) is referred to as Y0 and the new one is referred to as Y1]. A more detailed explanation of the model and the large-scale forcing data can be found in Wu et al. (2008).

To adopt the prescribed evolving surface albedo, at first the hourly observational albedo is calculated as the ratio of upward SW flux to downward SW flux at the surface. Since the CRM uses surface albedos of direct and diffuse incident radiation for two spectral intervals in the radiative transfer calculation, the broadband surface albedo (AB) is decomposed into two parts for wavelengths 0.2–0.7 μm (AS) and 0.7–5.0 μm (AL), respectively. When AB is equal to or smaller than 0.21, the ratios of AS/AB and AL/AB are set to 0.19 and 0.81, respectively (Briegleb 1992). When AB is equal to 1.0, AB is equally divided between AS and AL so that in the albedo range from 0.21 to 1.0, AS/AB and AL/AB are linearly increased and decreased, respectively, until AB = 1.0. The prescribed surface albedo is imposed in the computation of radiative fluxes and heating rates every 300 s. The difference of surface albedo between the two CRM runs is showed in Fig. 1 in terms of the seasonal change of diurnal variation. The albedo of Y0 is almost constant at 0.15 during daytime throughout four seasons, whereas that of Y1 has a distinct diurnal variation and also has relatively much greater values in winter. The surface albedo differences between Y0 and Y1 are about 0.05 in the spring, summer, and fall and 0.15 in the winter. Because of snow-covered periods, the standard deviations in winter are relatively larger than in other seasons. Usually early morning and late afternoon albedo is greater than during middaytime, and the early morning albedo is slightly greater than late afternoon albedo. This asymmetry is mainly caused by the direct beam albedo (Yang et al. 2008) and is possibly due to the dew effect in early morning (Minnis et al. 1997).

Fig. 1.

Seasonal-mean diurnal variation of surface albedo from Y0 and Y1; vertical solid bars indicate standard deviations from observations.

Fig. 1.

Seasonal-mean diurnal variation of surface albedo from Y0 and Y1; vertical solid bars indicate standard deviations from observations.

3. CRM-simulated radiative and cloud properties

The general characteristics of radiative fluxes and cloud properties from the CRM were shown by Wu et al. (2008) with constant surface albedo. In this section, the major effects of the prescribed evolving surface albedo on radiative and cloud properties are shown in annual and seasonal aspects.

The annual and seasonal means and standard deviations of net longwave (LW) and SW radiative fluxes at the TOA and the surface from Y1 and observations are listed in Table 1. The net flux is defined as downward minus upward fluxes. The observed TOA LW and SW fluxes are derived from the Geostationary Operational Environmental Satellite (GOES) (Minnis et al. 1995). The CRM-produced annual mean LW is close to the observed LW at the TOA with a difference less than 1 W m−2 and at the surface with a difference less than 6 W m−2. For the seasonal means, the differences of LW between Y1 and observations are within 7 W m−2 at the TOA and within 11 W m−2 at the surface. The differences of the annual mean SW flux between Y1 and observations are less than 8 W m−2 at the TOA and the surface. The use of prescribed surface albedo in Y1 makes wintertime SW flux much more comparable with observations. The difference from observations is about 3 W m−2 at the surface in winter. However, the discrepancies of SW annual radiative budgets compared to seasonal budgets between the CRM and observations are not negligible. The uncertainty in obtaining the area-mean surface SW flux from 22 stations may be partly responsible for those discrepancies (Li et al. 2002). The albedo difference in winter is important because there were many low-level clouds over the ARM SGP in the winter of 2000 (Wu et al. 2008). Table 2 lists annual and seasonal mean and standard deviation of upward and downward fluxes of SW and LW based on daily averaged values. At the surface, downward and upward LW budgets are very similar between Y1 and observations with differences of less than 2 W m−2 throughout the whole year, and upward SW flux budgets of Y1 are also very close to the observations owing to use of the prescribed evolving surface albedo. However, the differences of downward SW flux between Y1 and observations are large at the surface. In particular, upward SW fluxes at the surface from Y1 are close to observations. Annual means of Y1 and observations are 42.2 and 40.5 W m−2, and winter means are 35.3 and 32.4 W m−2, respectively. Since solar insolation at the TOA and the surface is usually much greater in summer than for other seasons, despite the relatively small albedo difference of 0.05 between Y0 and Y1, the corresponding effect on SW flux is very large. For example, the summer-mean solar insolation is 458 W m−2 and the winter mean is 214 W m−2 at TOA, and the mean surface albedo differences are 0.05 and 0.15 in summer and winter. However, the corresponding differences in net SW flux are about 10 W m−2 and 12 W m−2 in summer and winter, respectively. Therefore, surface albedo effect on the SW radiative budget depends on the intensity of solar insolation at the TOA throughout the entire year.

Table 1.

Annual (ANN) and seasonal (MAM, JJA, SON, and DJF) means and standard deviations (SD) of daily net (downward minus upward fluxes) LW and SW radiative fluxes (W m−2) at the TOA and surface (SFC) from Y1 and observations during the year 2000.

Annual (ANN) and seasonal (MAM, JJA, SON, and DJF) means and standard deviations (SD) of daily net (downward minus upward fluxes) LW and SW radiative fluxes (W m−2) at the TOA and surface (SFC) from Y1 and observations during the year 2000.
Annual (ANN) and seasonal (MAM, JJA, SON, and DJF) means and standard deviations (SD) of daily net (downward minus upward fluxes) LW and SW radiative fluxes (W m−2) at the TOA and surface (SFC) from Y1 and observations during the year 2000.
Table 2.

Annual and seasonal means and SD of dailySW upward (UP) and downward (DN) radiative fluxes (W m−2) at the TOA and SFC and LW fluxes at the SFC from Y1 and observations during the year 2000.

Annual and seasonal means and SD of dailySW upward (UP) and downward (DN) radiative fluxes (W m−2) at the TOA and SFC and LW fluxes at the SFC from Y1 and observations during the year 2000.
Annual and seasonal means and SD of dailySW upward (UP) and downward (DN) radiative fluxes (W m−2) at the TOA and SFC and LW fluxes at the SFC from Y1 and observations during the year 2000.

The primary difference in radiative flux between Y1 and Y0 is revealed in upward shortwave (SWUP) flux at the surface, as shown in Fig. 2. This figure indicates the daily variation of SWUP from Y0 and Y1 at the surface and differences from observations. Since the surface albedo of Y1 is usually greater than that of Y0, the SWUP of Y1 is generally greater than that of Y0 throughout the entire year. Especially, the winter values of Y1 are much larger when snow cover exists on the SGP site. The differences between Y0 and observations are usually large; the SWUP from Y0 is much smaller than that from observations in January, February, September, and December. The SWUP of Y1 is much improved during those periods.

Fig. 2.

Yearlong and diurnal variation of (top) upward SW flux (SWUP) at the surface (SFC) from Y0 and Y1 (top) based on hourly data and (bottom) the difference Y0 and Y1 from observations.

Fig. 2.

Yearlong and diurnal variation of (top) upward SW flux (SWUP) at the surface (SFC) from Y0 and Y1 (top) based on hourly data and (bottom) the difference Y0 and Y1 from observations.

Comparing Y1 with Y0 in terms of cloud properties such as liquid water path (LWP) and ice water path (IWP) gives insight as to the effect of surface albedo difference on cloud fields. The albedo effect on cloud systems might be an indirect process by which convection and temperature and moisture field could be varied after the change of radiative fluxes. Because of the prescribed surface heat fluxes, the surface albedo effects on cloud systems could be underestimated in this study. Figure 3 shows the seasonally averaged diurnal variation of vertically integrated LWP and IWP from Y0 and Y1 based on hourly averaged values. The differences between the two simulations are also shown. In the spring, Y1 has usually a greater amount of clouds at night—for instance, between 2000 and 0500 LST, about 5 and 20 g m−2 more in LWP and IWP, respectively—but fewer clouds just before sunset; daytime cloud amounts of Y1 are comparable with Y0. However, in the summer, Y1 usually has fewer clouds during the nighttime (1900–0300 LST, about 5 and 10 g m−2 less in LWP and IWP, respectively) and more clouds in the morning (0700–1100 LST). Y1 also has slightly more clouds before sunset in the summer. In the fall, Y1 cloud systems tend to have larger LWP (3 g m−2 more) and IWP (5 g m−2 more) than Y0 in the evening (1500–2300 LST). Throughout the entire day, Y1 clouds have about 2 g m−2 less LWP in the winter, but the amount of IWP of Y1 is almost comparable to Y0 except at 0300, 2000, and 2200 LST. The discrepancy between Y0 and Y1 in the winter is primarily caused by the great difference (0.15) in surface albedo between the two runs, which will be discussed in the next section.

Fig. 3.

Seasonal and diurnal variations of liquid water path (LWP) and ice water path (IWP) from Y0 and Y1, and the differences between the two simulations.

Fig. 3.

Seasonal and diurnal variations of liquid water path (LWP) and ice water path (IWP) from Y0 and Y1, and the differences between the two simulations.

4. Surface albedo effects

In addition to the annual and seasonal analyses of the year-long ISU CRM simulation with the prescribed evolving surface albedo in radiative fluxes and cloud properties, further analysis of the production from the CRM allows quantification of the effects of surface albedo on radiative fluxes and cloud properties.

a. Surface albedo effects on cloud properties and radiative fluxes

As shown in the previous section, the major surface albedo impact occursreveals usually in wintertime. From the CRM simulation, only 7 days are found as meaningful cases with large clouds (i.e., the sum of LWP and IWP is greater than 100 g m−2) and great surface albedos (i.e., greater than 0.4) in winter of the year 2000. However, most cases (six cases) have their major events at night and only one case occurs in daytime. From the case study during daytime, the relationship among surface albedo, radiative fluxes, and cloud properties can be investigated when surface albedo is very high and cloud amount is very large. Figure 4 shows the case during 27–29 January 2000 based on hourly mean values. Downward SW (SWDN) flux at the TOA (i.e., solar insolation) and upward longwave (LWUP) flux at the surface from Y1 must be the same as from Y0 because incoming solar radiation should be equivalent and the prescribed surface heat fluxes are same in the two simulations. During this period, the surface albedo of Y1 is much greater than for Y0 because the surface is covered by snow, and the difference of surface albedo between the two simulations is about 0.6. At the TOA, the SWUP of Y1 should be larger than that of Y0 as on the 28th and 29th owing to the surface albedo differences. However, the SWUP fluxes at the TOA on the 27th from both simulations are almost equivalent to each other in spite of the great surface albedo difference. That is primarily caused by clouds (Figs. 4h,i) reflecting SW flux from the cloud top. The SWDN fluxes at the surface are smaller on the 27th because of absorption by clouds, while the fluxes on the other days are large in the two simulations, as expected. However, SWUP at the surface from Y1 is much larger than that from Y0 owing to the great reflection from the surface. Considering Figs. 4d,e,h,i, it is noticed that SWDN at the surface can be affected by the cloud bottom reflection of SW flux coming from the surface. Therefore, the SW radiative flux is affected by surface albedo together with cloud albedo through the reflection at cloud top and base. Furthermore, in terms of LW flux (Figs. 4c,f), LWUP at the TOA and downward longwave (LWDN) flux at the surface are also different between the two simulations when clouds exist (Figs. 4h,i).

Fig. 4.

Radiative and cloud property snapshot for three days (27–29 Jan 2000) based on hourly data including SW and LW fluxes at the TOA and SFC for (a)–(g) downward (DN) and upward (UP) fluxes, (h) LWP, (i) IWP, and (j) surface albedo from Y0 and Y1.

Fig. 4.

Radiative and cloud property snapshot for three days (27–29 Jan 2000) based on hourly data including SW and LW fluxes at the TOA and SFC for (a)–(g) downward (DN) and upward (UP) fluxes, (h) LWP, (i) IWP, and (j) surface albedo from Y0 and Y1.

Since we are interested in the surface albedo effect on clouds, only daytime (i.e., 1300–2300 UTC) variations of the surface albedo and cloud properties are considered. During the daytime on 27th, IWP and LWP of Y1 are usually smaller than those of Y0 (the large IWP and LWP of Y1 in the early morning, before sunrise, will be discussed later). Unlike the variation of LWP in Y0, that of Y1 is dramatically decreased from the morning to late afternoon. A possible reason is that, because of greater surface albedo in the Y1 simulation, heating of the lower troposphere is increased due to additional absorption of SW radiation by water vapor and clouds so that temperature is increased (Fig. 5a). During daytime (1600 and 2100 UTC), the potential temperature of Y1 is larger than Y0 in the lower troposphere. The change of temperature profile leads to the change of temperature lapse rate. Figure 5b shows the vertical profiles of lapse rate (i.e., dT/dz) differences between the two simulations at 1200, 1600, and 2100 UTC. During daytime (1600 and 2100 UTC), the lapse rates in the lower troposphere are much smaller (−0.3° and −0.7°C km−1) in Y1, which leads to fewer clouds in Y1 simulation. Figure 6 illustrates the vertical profiles of cloud liquid and ice mixing ratios at two points in the daytime. The cloud graupel and rainwater mixing ratios are very small in this period. The amount of ice clouds is much less in Y1 and liquid clouds of Y1 also decrease from 1600 to 2100 UTC. The large amounts of IWP and LWP in the early morning could be explained by the greater lapse rate (+0.9°C km−1) in the lower troposphere before sunrise (Fig. 5, 1200 UTC). Since Y1 has fewer liquid and ice cloud particles during most of the daytime, the corresponding changes of radiative heating rates are also similar between 1600 and 2100 UTC (Fig. 6). There is less LW radiative cooling around cloud top, more cooling within clouds, and more cloud-base heating due to the weaker emission of LW flux from the surface, while there is more SW radiative cooling due to less reflection at the cloud top and more heating inside clouds because of more incoming SW radiation absorption. Thus, the large surface albedo induced fewer clouds, resulting in net radiative heating in the upper and lower troposphere and net radiative cooling in the middle troposphere in winter. Because of fewer clouds during daytime, the LWUP from the surface is less absorbed by the clouds, thereby slightly increasing LWUP at the TOA. Fewer clouds results in less LWDN from the cloud base, so LWDN at the surface is decreased. As shown in Figs. 4 –6, surface albedo affects not only the SW flux budget but also cloud systems via influencing temperature and instability in the lower troposphere, thereby affecting the LW flux as well. However, the surface albedo effect on cloud properties cannot be generalized with one case study. Usually in other cases in winter, the effect on clouds is small.

Fig. 5.

Vertical profiles of (a) potential temperature differences and (b) lapse rate differences between Y0 and Y1 at 1200, 1600, and 2100 UTC 27 Jan 2000.

Fig. 5.

Vertical profiles of (a) potential temperature differences and (b) lapse rate differences between Y0 and Y1 at 1200, 1600, and 2100 UTC 27 Jan 2000.

Fig. 6.

Vertical profiles of (left) domain-averaged cloud liquid and ice water mixing ratio from Y0 and Y1 and (right) radiative heating rate differences between Y0 and Y1 at (a) 1600 and (b) 2100 UTC.

Fig. 6.

Vertical profiles of (left) domain-averaged cloud liquid and ice water mixing ratio from Y0 and Y1 and (right) radiative heating rate differences between Y0 and Y1 at (a) 1600 and (b) 2100 UTC.

Cloud radiative forcing on SW and LW fluxes could be used to examine those processes among cloud, radiative fluxes, and surface albedo. In particular, the shortwave and longwave cloud radiative forcing (SWCF and LWCF, respectively) could generally be used as a measure of the large-scale effects of clouds on radiative fluxes. The cloud radiative forcing is defined as the difference between the net radiative flux of all sky and the net radiative flux corresponding to clear-sky conditions. In Table 3 seasonal mean values of SWCF and LWCF are obtained at the TOA and the surface for the Y0 and Y1 simulations are shown. The annual and seasonal mean values of CF from the two simulations averaged over the 600-km domain are listed. The annual and seasonal characteristics of CF from Y0 are well discussed by Wu et al. (2008). In terms of the annual mean, the two simulations of LWCF are almost equivalent to each other, whereas there are significant differences (e.g., 4 and 5 W m−2 at the TOA and the surface, respectively) between the simulations of SWCF. The discrepancies in SWCF between Y0 and Y1 are 2–4 W m−2 in the spring, summer, and fall at the TOA and the surface. However, the difference is significantly greater during wintertime, by 7–8 W m−2, primarily because of the different surface albedo. Moreover, in the seasonal aspect the LWCFs from the two simulations are almost the same during March–May (MAM), June–August (JJA), and September–November (SON), while the difference of winter LWCF at the surface between two simulations is much greater, by 13 W m−2, which might be caused by the frequently occurring low-level clouds over the ARM SGP in winter. To explicitly show the process between the surface albedo and near-surface clouds, surface temperature and surface heat fluxes need to physically interact with the surface albedo. Some implicit impacts are shown in this framework, but to achieve more realistic feedback of surface albedo to radiation and clouds, the CRM needs to be improved in its treatment of surface conditions.

Table 3.

Annual and seasonal means and SD of daily LW and SW cloud radiative forcing (W m−2) (all-sky minus clear-sky radiative fluxes) at the TOA and SFC from Y0 and Y1 during the year 2000.

Annual and seasonal means and SD of daily LW and SW cloud radiative forcing (W m−2) (all-sky minus clear-sky radiative fluxes) at the TOA and SFC from Y0 and Y1 during the year 2000.
Annual and seasonal means and SD of daily LW and SW cloud radiative forcing (W m−2) (all-sky minus clear-sky radiative fluxes) at the TOA and SFC from Y0 and Y1 during the year 2000.

The observed radiation and cloud data are generated from the variational analysis of NWP model-produced fields, including five ARM sounding locations, seven NOAA wind profiler locations, and the Rapid Update Cycle (RUC) analysis domain over the ARM SGP (Xie et al. 2004) so that cloud properties and surface albedo have the same large-scale temperature and moisture advective forcing, which facilitates investigating the relationship among surface albedo, cloud properties, and radiative fluxes. Figure 7 shows normalized SWUP flux at the TOA of Y1 with surface albedo and cloud water path (CWP) (i.e., LWP + IWP) variations based on daily mean values for the year 2000, indicating the relationship among the three variables (surface albedo, SWUP, and CWP). This figure gives a qualitative insight into the three variables. Since daily averaged quantities are used, the effect of the diurnal variation by SZA change does not need to be considered. When the CWP value is less than 400 g m−2, SWUP is slightly decreasing as surface albedo varies up to 0.25–0.35. When the albedo is greater than 0.35, as surface albedo is increasing to larger values, SWUP tends to increase as well. There is a critical value of surface albedo for affecting SWUP flux when CWP is relatively small. On the other hand, when CWP is relatively large, it seems that SWUP increases as surface albedo does without revealing any critical value of surface albedo. However, this is not clear in this analysis. It must be further examined whether or the increase of SWUP at the TOA is a result of the surface albedo increase when the CWP is large, because optically thick cloud usually has strong reflection at cloud top and base. Thus, there might be critical values of surface albedo and CWP that affect SW flux, which may support the fact that a more reflective surface (e.g., surface albedo >0.35) leads to weaker SWCF at the TOA for optically thin clouds, while less surface reflectivity (e.g., surface albedo <0.35) does not affect cloud forcing much. Also, when the surface albedo is less than the critical value, there is a weak decrease in SWUP at the TOA with increase of surface albedo, which suggests that a more reflective surface leads to weaker cloud forcing in thin clouds. Figure 7 also illustrates the relationship among surface albedo, CWP, and total cloud fraction. When CWP is relatively small (below 300 g m−2) and the surface albedo is large, the cloud fraction increases as the surface albedo increases, whereas this tendency is not clear when the surface albedo is small. It indicates that the increase of SWUP at the TOA with increasing surface albedo is partly affected by the increase of cloud fraction causing stronger reflection at cloud top when the surface albedo is large, although existing clouds are optically thin.

Fig. 7.

(left) Relationship among normalized SWUP flux at the TOA, surface albedo, and CWP (LWP + IWP) (left) and (right) the relationship with total cloud fraction, based on daily averaged values for the year 2000.

Fig. 7.

(left) Relationship among normalized SWUP flux at the TOA, surface albedo, and CWP (LWP + IWP) (left) and (right) the relationship with total cloud fraction, based on daily averaged values for the year 2000.

To specify the characteristics of the surface albedo effect revealed above, we further investigate how surface albedo is associated with SWUP at the TOA, considering specific ranges of LWP and SZA. Figure 8 is an example of the analysis and shows that all of the scattered spots in Fig. 8a can be decomposed in Figs. 8b and 8c using a critical value of LWP, that is, 50 g m−2 in this study. SWUP at the TOA is normalized by solar insolation at the TOA. Furthermore, Fig. 8b indicates a proportional relationship between surface albedo and SWUP at the TOA if surface albedo is greater than 0.35; otherwise there is less correlation between them. However, when LWP is greater than 50 g m−2 (i.e., optically thick clouds, Fig. 8c), SWUP at the TOA is almost constant, even as surface albedo increases from 0.35 to 0.7 and has less correlation when surface albedo is relatively small as well. It should be noticed that surface albedo affects SWUP at the TOA when cloud is optically thin (LWP < 50 g m−2) and surface albedo >0.35; otherwise, if cloud is optically thick or surface albedo is small, surface albedo does not have much effect on SWUP at the TOA. In addition, considering IWP with surface albedo and SWUP at the TOA also shows similar characteristics. The critical values of IWP and surface albedo are 150 g m−2 and 0.35. Consequently, all of the linear least squares fit lines are calculated for all SZA intervals considering the critical values of LWP and IWP in Fig. 9. As shown in Fig. 8, there are two regression lines for each SZA: one for albedos from 0.1 to 0.35 and the other for 0.35 to 0.85. There are obviously proportional relationships between surface albedo and SWUP at the TOA when surface albedo is larger than 0.35 and LWP is smaller than 50 g m−2 (Fig. 9a). Besides, as SZA increases, the slopes of regression lines decrease. However, when surface albedo is small or LWP is large, there is weak correlation between surface albedo and SWUP at the TOA. IWP cases also have features similar to the LWP cases. Therefore, surface albedo strongly influences SWUP at the TOA if cloud is optically thin (i.e., LWP and IWP < 50 and 150 g m−2, respectively), surface albedo is greater than 0.35, and SZA small. In other words, clouds do not greatly affect the positively proportional relationship between surface albedo and SWUP at the TOA when the existing clouds are optically thin, surface albedo is large, and SZA is small. However, if cloud is optically thick, surface albedo small, or SZA large, the surface albedo impact on SWUP at the TOA could be suppressed.

Fig. 8.

Scatter diagrams of normalized SWUP flux at the TOA vs surface albedo for (a) all LWP ranges, (b) below 50 g m−2, and (c) above 50 g m−2 when the SZA is between 60° and 70°.

Fig. 8.

Scatter diagrams of normalized SWUP flux at the TOA vs surface albedo for (a) all LWP ranges, (b) below 50 g m−2, and (c) above 50 g m−2 when the SZA is between 60° and 70°.

Fig. 9.

Distributions of linear least squares fit regression lines for the relationship between normalized SWUP flux at the TOA and surface albedo for (a) small LWP, (b) large LWP, (c) small IWP, and (d) large IWP cases within each SZA range. There are two regression lines for each SZA range, representing albedos <0.35 and >0.35.

Fig. 9.

Distributions of linear least squares fit regression lines for the relationship between normalized SWUP flux at the TOA and surface albedo for (a) small LWP, (b) large LWP, (c) small IWP, and (d) large IWP cases within each SZA range. There are two regression lines for each SZA range, representing albedos <0.35 and >0.35.

Furthermore, at the surface SWUP is directly proportional to surface albedo (not shown) in all ranges of LWP and IWP, as expected, without any critical values of surface albedo, LWP, and IWP. However, SWDN at the surface has characteristics similar to SWUP at the TOA (Fig. 10). SWDN at the surface is more associated with CWP than with LWP and IWP. This might be caused by the fact that clouds near the surface usually comprise both liquid and ice particles except in summer. Similar to SWUP at the TOA, SWDN at the surface is normalized by solar insolation at the TOA. SWDN at the surface has a relatively smaller correlation with surface albedo when the albedo is less than 0.35. Especially when CWP is greater than 180 g m−2, SWDN at the surface has only weak linear correlation with surface albedo larger than 0.35; SWDN at the surface is proportional to surface albedo as well. The correlation between them decreases as SZA increases. This result is consistent with a high-latitude case study by Nichol et al. (2003). They explained that a great amount of surface albedo can moderate the attenuation of SW radiation by cloud through multiple scattering between cloud base and the surface. However, this phenomenon takes place only if optically thick cloud exists, that is, CWP > 180 g m−2); otherwise, SWDN at the surface is not varied as surface albedo increases when cloud is optically thin (not shown). Thus, the reflection at cloud base could be another reason for that.

Fig. 10.

Scatter diagrams of normalized SWDN at the SFC vs surface albedo for each SZA range. Solid lines are linear least squares fit regression lines considering surface albedos greater than 0.35.

Fig. 10.

Scatter diagrams of normalized SWDN at the SFC vs surface albedo for each SZA range. Solid lines are linear least squares fit regression lines considering surface albedos greater than 0.35.

Therefore, when optically thick cloud exists, SWUP at the TOA is primarily associated with reflection at cloud top and is not much affected by the surface albedo, while the albedo influences SWDN at the surface by cloud-base reflection and multiple scattering between cloud base and the surface. In other words, clouds have a negligible feedback on SWUP at the TOA and slightly positive feedback on SWDN at the surface in association with surface albedo if the cloud is optically thick. On the other hand, when optically thin cloud occurs, SWUP at the TOA is mainly influenced by the surface albedo rather than cloud-top reflection, while SWDN at the surface is not affected by the change. SWUP at the surface is generally strongly affected by surface albedo for all cloud cases, as expected.

b. Solar zenith angle impact on the surface albedo effect

Surface albedo directly affects SW flux and SWCF and indirectly affects cloud systems, thereby influencing LW flux and LWCF also. In particular, surface albedo strongly influences SW radiative flux when SZA is small. Further analysis could be useful to quantify how those surface albedo effects depend on SZA.

As illustrated in Fig. 1, the surface albedo in the Y1 simulation has clear diurnal and seasonal variation. To understand the variation quantitatively, Fig. 11 shows the seasonal variation of scatter diagrams between surface albedo and cosine of SZA from Y1. Throughout all seasons, the surface albedo usually tends to have smaller values with small SZA than with large SZA, similar to the illustration in Fig. 1. The numbers on each scatter diagram indicate the value of a slope from linear least squares fit calculation. The slope is relatively large in winter, revealing a great dependence of the surface albedo on SZA variation, mainly due to snow melting during daytime.

Fig. 11.

Scatter diagrams of surface albedo vs cos(SZA) for each season from Y1. Numbers for each season indicate the slope of linear least squares fit regression.

Fig. 11.

Scatter diagrams of surface albedo vs cos(SZA) for each season from Y1. Numbers for each season indicate the slope of linear least squares fit regression.

As shown in Figs. 9 and 10, the slope from a linear least squares fit regression between surface albedo and radiative fluxes could be a good parameter to quantify how the surface albedo effect on radiation depends on SZA. As revealed in the previous section, upward SW flux at the TOA is significantly affected by surface albedo when optically thin clouds exist, whereas downward SW flux at the surface is influenced by albedo with optically thick clouds. Also, upward SW flux at the surface is always affected by the albedo, regardless of the existing cloud optical thickness, as expected. Figure 12 illustrates normalized slopes as a function of SZA for each radiative flux. The slopes from a linear least squares fit regression between surface albedo and radiative flux for each SZA bin (10°) are normalized by cosine of SZA and the slope when SZA is 50°. The surface albedo effect on upward SW flux at the TOA for optically thin clouds (i.e., LWP < 50 g m−2) is greater when SZA is 50°, 60°, and 70° and smaller at SZA 80° and 90° (Fig. 12a). The effect on downward SW flux at the surface with optically thick clouds (i.e., CWP ≥ 180 g m−2) is much greater at SZA = 50° but smaller at other SZAs (Fig. 12b), which probably indicates that SZA influences not only the surface albedo effect on downward SW flux at the surface but also the reflection of SW flux from the cloud base. The surface albedo impact on upward SW flux at the surface almost linearly decreases as SZA increases. At the surface, the slope is significantly decreased for 50° to 60° SZA, while the same change of slope at the TOA is relatively small. Therefore, the SZA change affects the SW radiative budget more at the surface than at the TOA. This analysis might have limitations generalizing the result. Since large surface albedo cases occur only in winter, the sample size may not be enough even though yearlong data from the CRM is considered.

Fig. 12.

Normalized slopes as a function of SZA for (a) SWUP flux at the TOA for optically thin clouds (LWP < 50 g m−2), (b) SWDN flux at the SFC for optically thick clouds (CWP ≥ 180 g m−2), and (c) SWUP flux at the surface for all clouds.

Fig. 12.

Normalized slopes as a function of SZA for (a) SWUP flux at the TOA for optically thin clouds (LWP < 50 g m−2), (b) SWDN flux at the SFC for optically thick clouds (CWP ≥ 180 g m−2), and (c) SWUP flux at the surface for all clouds.

5. Summary and discussion

A yearlong ISU CRM simulation was conducted with prescribed evolving surface albedo to investigate the relationship among surface albedo, radiative fluxes, and cloud properties. The CRM simulation well represents the shortwave radiative budget during winter because the radiation calculation for the snow-covered period is improved by using the prescribed evolving surface albedo. The surface albedo effect on the shortwave radiative budget depends on the intensity of solar insolation at the TOA throughout the entire year, and the relative effect increases as the shortwave intensity increases. Shortwave cloud radiative forcing at the surface and TOA is decreased by the use of evolving surface albedo throughout the whole year and more so in winter. However, the longwave cloud radiative forcing at the surface and TOA are not much influenced by the evolving surface albedo except in winter.

Shortwave radiative flux at the surface and TOA are affected by the evolving surface albedo and clouds through the reflection at cloud top and base. Longwave flux at the surface and TOA are also affected by the evolving surface albedo through its feedback with clouds. In winter, the larger surface albedo causes more heating of the lower troposphere due to additional absorption of shortwave radiation by water vapor and clouds, which reduces the temperature lapse rate. The weaker instability in daytime causes fewer clouds in the lower troposphere. The change of cloud properties due to the evolving surface albedo leads to a change of the longwave radiative budget at the surface and TOA also. However, the effect on cloud properties is small in terms of annual and seasonal means.

Moreover, it is found that 0.35 is a critical value of surface albedo that affects the upward shortwave flux at the TOA and downward shortwave flux at the surface for optically thin and thick clouds, respectively. When the existing clouds are optically thin (e.g., LWP < 50 g m−2 and IWP < 150 g m−2), upward shortwave flux at the TOA has a proportional relationship with surface albedos greater than the critical value. Downward shortwave flux at the surface also has a proportional relationship with surface albedos larger than the critical values if optically thick (e.g., CWP > 180 g m−2) clouds exist. The relationship between the surface albedo and shortwave flux has a dependence on SZA; the surface albedo effect is greater for smaller SZAs, but the effect is almost equivalent at higher SZAs. However, the dependence on SZA is negligible when LWP and IWP are large enough. In general, therefore, when optically thick cloud exists, the upward shortwave flux at the TOA is primarily influenced by reflection at cloud top but is not much affected by the surface albedo, whereas the albedo influences the downward shortwave flux at the surface by cloud-base reflection and multiple scattering between cloud base and the surface. In other words, cloud has a negligible feedback on the upward shortwave flux at the TOA and slightly positive feedback on the downward shortwave flux at the surface in association with surface albedo if the cloud is optically thick. On the other hand, when optically thin cloud occurs, the upward shortwave flux at the TOA is mainly influenced by the surface albedo rather than the cloud-top reflection, while the downward shortwave flux at the surface is not affected by the surface albedo.

Surface albedos have a slightly proportional relationship with SZAs in the CRM. The dependence of surface albedo on SZA is greater in winter than in other seasons. The SZA change affects the shortwave radiation budget at the surface more than at the TOA. Shortwave cloud radiative forcing is also affected by SZA: as SZA decreases, shortwave cloud radiative forcing increases.

One limitation of this study is that the surface albedo effects on cloud properties might be underestimated owing to the use of prescribed evolving surface sensible and latent heat fluxes, which probably results in the underestimation of the surface albedo impact on the longwave radiation. If a land surface model is coupled with the CRM, more physically coherent effects of surface albedo on cloud and radiative properties would be considered and quantified.

Acknowledgments

We thank numerous ARM science team members for producing the ARM forcing data and observational estimates to conduct and evaluate CRM simulations. Special thanks go to Dr. Qilong Min for discussing the use of observed surface albedo. This research was partly supported by the Biological and Environmental Research Program (BER), the U.S. Department of Energy under Grant DE-FG02-08ER64559, and the National Sciences Foundation under Grant ATM-0935263. Computing support by Daryl Herzmann is greatly appreciated.

REFERENCES

REFERENCES
Baker
,
D. G.
, and
D. L.
Ruschy
,
1989
:
Winter albedo characteristics at St. Paul, Minnesota.
J. Appl. Meteor.
,
28
,
227
232
.
Briegleb
,
B. P.
,
1992
:
Delta-Eddington approximation for solar radiation in the NCAR Community Climate Model.
J. Geophys. Res.
,
97
,
7603
7612
.
Clark
,
T. L.
,
W. D.
Hall
, and
J. L.
Coen
,
1996
:
Source code documentation for the Clark–Hall cloud-scale model: Code version G3CH01.
NCAR Tech. Note NCAR/TN-426+STR, 137 pp
.
Dong
,
X.
,
2005
:
The impact of surface albedo on the retrievals of low-level stratus cloud properties: An updated parameterization.
Geophys. Res. Lett.
,
32
,
L10802
.
doi:10.1029/2005GL022548
.
Duchon
,
C. E.
, and
K. G.
Hamm
,
2006
:
Broadband albedo observations in the Southern Great Plains.
J. Appl. Meteor. Climatol.
,
45
,
210
235
.
Grabowski
,
W. W.
,
X.
Wu
, and
M. W.
Moncrieff
,
1996
:
Cloud resolving modeling of tropical cloud systems during Phase III of GATE. Part I: Two-dimensional experiments.
J. Atmos. Sci.
,
53
,
3684
3709
.
Grant
,
I. F.
,
A. J.
Prata
, and
R. P.
Cechet
,
2000
:
The impact of the diurnal variation of albedo on the remote sensing of the daily mean albedo of grassland.
J. Appl. Meteor.
,
39
,
231
244
.
Hall
,
A.
,
2004
:
The role of surface albedo feedback in climate.
J. Climate
,
17
,
1550
1568
.
Holtslag
,
A. A. M.
, and
C-H.
Moeng
,
1991
:
Eddy diffusivity and countergradient transport in the convective atmospheric boundary layer.
J. Atmos. Sci.
,
48
,
1690
1698
.
Hong
,
S-Y.
, and
H-L.
Pan
,
1996
:
Nonlocal boundary layer vertical diffusion in a medium-range forecast model.
Mon. Wea. Rev.
,
124
,
2322
2339
.
Huber
,
M.
,
M.
Blumthaler
,
J.
Schreder
,
B.
Schallhart
, and
J.
Lenoble
,
2004
:
Effect of inhomogeneous surface albedo on diffuse UV sky radiance at a high-altitude site.
J. Geophys. Res.
,
109
,
D08107
.
doi:10.1029/2003JD004013
.
Kessler
,
E.
,
1969
:
On the Distribution and Continuity of Water Substance in Atmospheric Circulations.
Meteor. Monogr., No. 32, Amer. Meteor. Soc., 84 pp
.
Kiehl
,
J. T.
,
J. J.
Hack
,
G. B.
Bonan
,
B. A.
Boville
,
B. P.
Briegleb
,
D. L.
Williamson
, and
P. J.
Rasch
,
1996
:
Description of the NCAR Community Climate Model (CCM3).
NCAR Tech. Note, NCAR/TN420+STR, 152 pp. [Available from NCAR Information Service, P.O. Box 3000, Boulder, CO 80307]
.
Koenig
,
L. R.
, and
F. W.
Murray
,
1976
:
Ice-bearing cumulus cloud evolution: Numerical simulation and general comparison against observations.
J. Appl. Meteor.
,
15
,
747
762
.
Kylling
,
A.
,
A.
Dahlback
, and
B.
Mayer
,
2000
:
The effect of clouds and surface albedo on UV irradiances at a high latitude site.
Geophys. Res. Lett.
,
27
,
1411
1414
.
Li
,
Z.
,
M. C.
Cribb
, and
A. P.
Trischenko
,
2002
:
Impact of surface inhomogeneity on solar radiative transfer under overcast conditions.
J. Geophys. Res.
,
107
,
4294
.
doi:10.1029/2001JD000976
.
McKenzie
,
R. L.
,
K. J.
Paulin
, and
S.
Madronich
,
1998
:
Effects of snow cover on UV irradiance and surface albedo: A case study.
J. Geophys. Res.
,
103
,
28785
28792
.
Minnis
,
P.
,
W. L.
Smith
Jr.
,
D. P.
Garber
,
J. K.
Ayers
, and
D. R.
Doelling
,
1995
:
Cloud properties derived from GOES-7 for spring 1994 ARM intensive observing period using version 1.0.0 of ARM satellite data analysis program.
NASA Reference Publication 1366, 59 pp
.
Minnis
,
P.
,
S.
Mayor
,
W. L.
Smith
Jr.
, and
D. F.
Young
,
1997
:
Asymmetry in the diurnal variation of surface albedo.
IEEE Trans. Geosci. Remote Sens.
,
35
,
879
891
.
Nichol
,
S.
,
G.
Pfister
,
G.
Bodeker
,
R.
McKenzie
,
S.
Wood
, and
G.
Bernhard
,
2003
:
Moderation of cloud reduction of UV in the Antarctic due to high surface albedo.
J. Appl. Meteor.
,
42
,
1174
1183
.
Qu
,
X.
, and
A.
Hall
,
2007
:
What controls the strength of snow–albedo feedback?
J. Climate
,
20
,
3971
3981
.
Randall
,
D. A.
, and
Coauthors
,
1994
:
Analysis of snow feedbacks in 14 general circulation models.
J. Geophys. Res.
,
99
,
20757
20772
.
Schneider
,
S. H.
, and
R. E.
Dickinson
,
1974
:
Climate modeling.
Rev. Geophys. Space Phys.
,
12
,
447
493
.
Shupe
,
M. D.
, and
J. M.
Intrieri
,
2004
:
Cloud radiative forcing of the arctic surface: The influence of cloud properties, surface albedo, and solar zenith angle.
J. Climate
,
17
,
616
628
.
Smagorinsky
,
J.
,
1963
:
General circulation experiments with the primitive equations. I. The basic experiment.
Mon. Wea. Rev.
,
91
,
99
164
.
Stamnes
,
K.
,
J.
Slusser
,
M.
Bowen
,
C.
Booth
, and
T.
Lucas
,
1990
:
Biologically effective ultraviolet radiation, total ozone abundance, and cloud optical depth at McMurdo Station, Antarctica, September 15 1988 through April 15 1989.
Geophys. Res. Lett.
,
17
,
2181
2184
.
Troen
,
I.
, and
L.
Mahrt
,
1986
:
A simple model of the atmospheric boundary layer: Sensitivity to surface evaporation.
Bound.-Layer Meteor.
,
37
,
129
148
.
Tsvetsinskaya
,
E. A.
,
C. B.
Schaaf
,
F.
Gao
,
A. H.
Strahler
,
R. E.
Dickinson
,
X.
Zeng
, and
W.
Lucht
,
2002
:
Relating MODIS-derived surface albedo to soils and rock types over Northern Africa and the Arabian peninsula.
Geophys. Res. Lett.
,
29
,
1353
.
doi:10.1029/2001GL014096
.
Wang
,
Z.
,
M.
Barlage
,
X.
Zeng
,
R. E.
Dickinson
, and
C. B.
Schaaf
,
2005
:
The solar zenith angle dependence of desert albedo.
Geophys. Res. Lett.
,
32
,
L05403
.
doi:10.1029/2004GL021835
.
Winton
,
M.
,
2006
:
Surface albedo feedback estimates for the AR4 climate models.
J. Climate
,
19
,
359
365
.
Wu
,
X.
, and
M. W.
Moncrieff
,
2001
:
Long-term behavior of cloud systems in TOGA COARE and their interactions with radiative and surface processes. Part III: Effects on the energy budget and SST.
J. Atmos. Sci.
,
58
,
1155
1168
.
Wu
,
X.
,
W. W.
Grabowski
, and
M. W.
Moncrieff
,
1998
:
Long-term behavior of cloud systems in TOGA COARE and their interactions with radiative and surface processes. Part I: Two-dimensional modeling study.
J. Atmos. Sci.
,
55
,
2693
2714
.
Wu
,
X.
,
W. D.
Hall
,
W. W.
Grabowski
,
M. W.
Moncrieff
,
W. D.
Collins
, and
J. T.
Kiehl
,
1999
:
Long-term behavior of cloud systems in TOGA COARE and their interactions with radiative and surface processes. Part II: Effects of ice microphysics on cloud–radiation interaction.
J. Atmos. Sci.
,
56
,
3177
3195
.
Wu
,
X.
,
X-Z.
Liang
, and
S.
Park
,
2007
:
Cloud-resolving model simulations over the ARM SGP.
Mon. Wea. Rev.
,
135
,
2841
2853
.
Wu
,
X.
,
S.
Park
, and
Q.
Min
,
2008
:
Seasonal variation of cloud systems over ARM SGP.
J. Atmos. Sci.
,
65
,
2107
2129
.
Xie
,
S.
,
R. T.
Cederwall
, and
M.
Zhang
,
2004
:
Developing long-term single-column model/cloud system-resolving model forcing data using numerical weather prediction products constrained by surface and top of the atmosphere observations.
J. Geophys. Res.
,
109
,
D01104
.
doi:10.1029/2003JD004045
.
Yang
,
F.
,
H-L.
Pan
,
S. K.
Krueger
,
S.
Moorthi
, and
S. J.
Lord
,
2006
:
Evaluation of the NCEP Global Forecast System at the ARM SGP site.
Mon. Wea. Rev.
,
134
,
3668
3690
.
Yang
,
F.
,
K.
Mitchell
,
Y-T.
Hou
,
Y.
Dai
,
X.
Zeng
,
Z.
Wang
, and
X-Z.
Liang
,
2008
:
Dependence of land surface albedo on solar zenith angle: Observations and model parameterization.
J. Climate Appl. Meteor.
,
47
,
2963
2982
.

Footnotes

Corresponding author address: Sunwook Park, Iowa State University, 3010 Agronomy Hall, Ames, IA 50011. Email: wsunwook@iastate.edu