15 MARCH 2015
SEIKI ET AL.
2405
Improvement in Global Cloud-System-Resolving Simulations by Using a Double-Moment Bulk Cloud Microphysics Scheme TATSUYA SEIKI, CHIHIRO KODAMA, AND AKIRA T. NODA Japan Agency for Marine-Earth Science and Technology, Yokohama, Japan
MASAKI SATOH Japan Agency for Marine-Earth Science and Technology, Yokohama, and Atmosphere and Ocean Research Institute, The University of Tokyo, Tokyo, Japan (Manuscript received 30 March 2014, in final form 27 November 2014) ABSTRACT This study examines the impact of an alteration of a cloud microphysics scheme on the representation of longwave cloud radiative forcing (LWCRF) and its impact on the atmosphere in global cloud-systemresolving simulations. A new double-moment bulk cloud microphysics scheme is used, and the simulated results are compared with those of a previous study. It is demonstrated that improvements within the new cloud microphysics scheme have the potential to substantially improve climate simulations. The new cloud microphysics scheme represents a realistic spatial distribution of the cloud fraction and LWCRF, particularly near the tropopause. The improvement in the cirrus cloud-top height by the new cloud microphysics scheme substantially reduces the warm bias in atmospheric temperature from the previous simulation via LWCRF by the cirrus clouds. The conversion rate of cloud ice to snow and gravitational sedimentation of cloud ice are the most important parameters for determining the strength of the radiative heating near the tropopause and its impact on atmospheric temperature.
1. Introduction After the work by Cess et al. (1989), the importance of cloud feedbacks for climate projection has been revealed using general circulation models (GCMs); cloud feedbacks are the major cause of significant diversity in simulated results. The elaboration of cloud processes can lead to improvements in climate projection; however, GCMs necessarily assume subgrid cloud processes because of their coarse spatiotemporal resolution. The subgrid cloud processes are highly parameterized; the modeling of those processes has been an issue over the last few decades. Another way to contribute to understanding
Denotes Open Access content.
Corresponding author address: Tatsuya Seiki, Project Team for Risk Information on Climate Change, Japan Agency for MarineEarth Science and Technology, 3173-15, Showa-machi, Kanazawa-ku, Yokohama, Kanagawa 236-0001, Japan. E-mail:
[email protected] DOI: 10.1175/JCLI-D-14-00241.1 Ó 2015 American Meteorological Society
cloud processes is global high-resolution simulations using detailed cloud microphysics schemes. Recently, global cloud-system-resolving models (GCRMs) and multiscale modeling frameworks (MMFs) have been developed to investigate the roles of convective cloud systems in large-scale atmospheric disturbances and general circulation using cloud microphysics schemes without cumulus parameterizations (Grabowski 2001; Khairoutdinov and Randall 2003; Tomita and Satoh 2004; Khairoutdinov et al. 2008; Satoh et al. 2008; Tao et al. 2009). Although GCRMs and MMFs have been successful in exploiting simulations of multiscale interactions over the tropics, simulation biases exist in precipitation, cloud amount, and cloud radiative forcing (CRF) (Khairoutdinov et al. 2008; Tao et al. 2009; Kodama et al. 2012; Hashino et al. 2013). It is expected that these systematic biases associated with hydrological cycles will be reduced by improving the representation of cloud microphysical processes. The Nonhydrostatic Icosahedral Atmospheric Model (NICAM ; Tomita and Satoh 2004; Satoh et al. 2008, 2014) GCRM uses the single-moment bulk cloud microphysics
2406
JOURNAL OF CLIMATE
schemes proposed by Grabowski (1998) and Tomita (2008). Because global simulations with a horizontal resolution of a few kilometers are quite expensive despite of the use of massively parallel computers, the aforementioned simple cloud microphysics schemes are acceptable (e.g., the computational cost of the simulations using a double-moment bulk cloud microphysics scheme described here is approximately twice as large as that using the aforementioned simple cloud microphysics schemes). Only a limited number of studies have used a double-moment bulk cloud microphysics scheme in either MMFs or GCRMs (Wang et al. 2011a,b). Several studies have shown that the cloud optical properties, radiative budget, and structure of convective clouds and associated anvil clouds are sensitive to the assumed values of empirical parameters used in such simple cloud microphysics schemes (Iga et al. 2007; Satoh et al. 2010; Kodama et al. 2012; Hashino et al. 2013). For general use in climate simulations such as climate change conditions, use of empirical formulations with uncertain parameters in cloud microphysics schemes must be avoided. In general, cloud microphysics schemes used in regional nonhydrostatic models have not been able to be utilized for global simulations because their numerical results have been typically evaluated in specific local regions (e.g., Ikawa et al. 1991). Recently, various observational campaigns and satellite measurements have provided the modeling community with extensive datasets, and then cloud microphysics schemes have been statistically evaluated and improved for the purpose of general use based on these datasets (e.g., Matsui et al. 2009; Lang et al. 2011; 2014; Tao et al. 2014). In particular, general information about hydrometeor particle size distributions (PSDs) can be derived from an analysis of various observational datasets; methods to diagnose PSDs have contributed to the improvements of singlemoment bulk cloud microphysics schemes (Field et al. 2005, 2007; Thompson et al. 2008; Abel and Boutle 2012). The diagnosis methods for PSDs have been tested against various types of cloud systems, and their effectiveness to reproduce cloud systems has also been reported (e.g., Lang et al. 2011). Recently, Roh and Satoh (2014) compiled a method to diagnose PSDs and showed improvements in a single-moment bulk cloud microphysics scheme used in NICAM (Tomita 2008; i.e., the NICAM singlemoment scheme with six water categories, which is referred to as NSW6). It is promising that the realistic representation of hydrometeor PSDs in numerical models will improve the performance of global cloud-systemresolving simulations. However, the cloud microphysical processes in NSW6 scheme remain the simple formulations that are used in conventional regional nonhydrostatic models and will be improved. Our objective is to examine
VOLUME 28
the change in the global distribution of simulated clouds by replacing the NSW6 scheme with a cloud microphysics scheme that includes different representations of microphysical processes and to investigate the reasons for the differences in the simulated results. In parallel with the development of single-moment bulk microphysics schemes used in GCRMs, Seiki and Nakajima (2014) developed a double-moment bulk cloud microphysics scheme based on Seifert and Beheng (2006) for NICAM (i.e., the NICAM double-moment scheme with six water categories, which is referred to as NDW6). Double-moment bulk cloud microphysics schemes represent PSDs using the number concentration and mixing ratio of hydrometeors as the prognostic variables; the mode radii are dynamically evaluated. The mode radii of hydrometeors are particularly important for the initiation of precipitating particles through collisional processes (Berry and Reinhardt 1974; Seifert and Beheng 2001). After the initiation of precipitation, gravitational sedimentation further distorts the PSD, and the time evolution of the PSD affects the subsequent chain of particle growth (Wacker and Seifert 2001; Milbrandt and Yau 2005; Milbrandt and McTaggart-Cowan 2010). The NSW6 scheme uses empirical formulations with tuned parameters in these collisional and sedimentation processes. These processes are validated by only the radiation budget at the top of the atmosphere and are problematic from the aspect of the vertical structure of clouds (Kodama et al. 2012). In contrast, Seifert’s scheme has been well evaluated compared with a spectral bin cloud microphysics scheme and observations (Seifert and Beheng 2001; Seifert et al. 2006; Seifert 2008). We expect that a comparison of the NDW6 and NSW6 schemes will provide insight into the improvements of cloud microphysics schemes in GCRMs and MMFs. In addition, some GCMs have begun to implement detailed cloud microphysics schemes (e.g., Gettelman et al. 2008; Morrison and Gettelman 2008; Salzmann et al. 2010). This study will be increasingly informative for long-term simulations using such GCMs. In this study, we concentrate on the change in simulated cloud ice. We compare the distributions of solid hydrometeors and longwave cloud radiative forcing (LWCRF) among simulated results using the NDW6 and NSW6 schemes and retrieved datasets from satellite observations. We then examine the change in the impact of cirrus cloud radiative forcing on atmospheric temperature using the NDW6 scheme. In section 2, we describe the experimental design of the global simulations. In section 3, we introduce the observational data used for comparisons. Moreover, in section 4, we present the results. We discuss the origin of biases in the atmospheric
15 MARCH 2015
2407
SEIKI ET AL.
temperature in section 5. In section 6, the conclusions are presented.
2. Experimental designs a. Control experiment As a reference, the latest simulated results of the control run in the study by Kodama et al. (2012) are used herein (hereafter referred to as CTL). In the simulation, NICAM was used as a GCRM. The horizontal grid interval was approximately 14 km over the globe. The model top was set to 40-km altitude, and the vertical model domain had 40 layers. The authors analyzed 3-month simulations beginning on 1 June 2004. Recently, it has been revealed that systematic biases in physical processes (e.g., cloud microphysics) in climate simulations using GCMs appear in short-range forecasts (Phillips et al. 2004; Martin et al. 2010; Williams et al. 2013). Thus, it is expected that 3-month simulations will provide characteristic differences in the performance of cloud microphysics schemes. Cumulus parameterizations are avoided, and cloud growth is explicitly calculated in each grid box using the NSW6 scheme. The NSW6 scheme predicts the mixing ratio of cloud water, rain, cloud ice, snow, and graupel (qc, qr, qi, qs, and qg, respectively), and its microphysical formulations are based on Lin et al. (1983) and Rutledge and Hobbs (1983). Radiative transfer is calculated using a broadband model with 29 spectral bands (Sekiguchi and Nakajima 2008). The cloud optical properties are calculated in advance according to Mie theory, in which the refractive indices between liquid and solid hydrometeors are separated, and determined based on lookup tables as a function of the effective radii (Nakajima et al. 2000). In the CTL experiment, the effective radii of the liquid and solid hydrometeors are assumed to be 8 and 40 mm, respectively. The numerical settings of other physical processes are briefly summarized here; Kodama et al. (2012) should be referred to for additional details. The subgrid turbulent mixing is calculated using level 2 of the second-order closure scheme with moist processes (Nakanishi and Niino 2004; Noda et al. 2010). The land surface model is the Minimal Advanced Treatments of Surface Interaction and Runoff (MATSIRO; Takata et al. 2003), and a single-layer slab ocean model is used. The bulk surface fluxes over the ocean are calculated using the method proposed by Louis (1979). The initial conditions are determined by linearly interpolating the NCEP final operational grid analysis (FNL) ds083.2 product. Kodama et al. (2012) reported that the CTL experiment captures the characteristic signal of deep convective clouds, which is a boomerang pattern in the radar
reflectivity–height histogram, according to the CloudSat satellite observations, whereas the cirrus cloud fraction is overestimated compared to the CALIPSO data. The empirical parameters in the NSW6 scheme in the CTL experiment are selected to adjust the radiation budget at the top of atmosphere: sedimentation of the cloud ice is switched off and the critical value of the mixing ratio (qicrt) for autoconversion from cloud ice to snow is set to 5 mg kg21 (see the appendix).
b. New cloud microphysics scheme The double-moment bulk cloud microphysics scheme, NDW6, predicts the number concentration of cloud water, rain, cloud ice, snow, and graupel (Nc, Nr, Ni, Ns, and Ng, respectively) in addition to their mixing ratios. We use the number mixing ratio, which is the number concentration divided by the air density, in advection, sedimentation, and turbulent mixing processes. The merits of the use of the number mixing ratio instead of the number concentration in these processes were argued by Mansell (2010). Cloud microphysical processes in the scheme are based on Seifert and Beheng (2001, 2006) and Seifert (2008), and the differences from the original scheme are described in detail by Seiki and Nakajima (2014) and Seiki et al. (2014). Here, we briefly describe the numerical modeling of cloud ice. Nonsphericity of ice hydrometeors is characterized by power-law relationships between the particle mass x and maximum dimension D and between the particle mass and projected area perpendicular to the flow A: D 5 am xbm
and
(1)
A 5 aA xbA ,
(2)
where am, bm, aA, and bA are constant coefficients. We assume that cloud ice can be represented by hexagonal columns, snow as assemblages of polycrystals, and graupel as lump graupel; the coefficient values are those compiled by Mitchell (1996). In the NDW6 scheme, the power-law relationships are used for calculation of the terminal velocity, the collisional cross section is used in the aggregation process, and the capacitance is used in the deposition and sublimation processes. The scheme is coupled to the radiative transfer model for calculating the cloud optical properties from a lookup table of the effective radii re. Following Fu (1996), the effective radius is defined as the ratio of the ice water content to the total projected area perpendicular to the flow as follows: r e, j [
3 ð 4ri ‘ 0
rqj Aj (xj )fj (xj ) dxj
for
j 5 i, s, g,
(3)
2408
JOURNAL OF CLIMATE
where ri 5 916.7 kg m23 is the density of ice, r is the air density, and f(x) is a particle mass distribution function. Using broad lookup table that covers the effective radii from 1 mm to 1 mm, the scheme can estimate the CRF including precipitating hydrometeors. In addition, we assume nonspherical single scattering properties for ice hydrometeors based on Fu (1996) and Fu et al. (1998). This is the first study to utilize the NDW6 scheme in a global simulation. Aerosol transport models have not yet been coupled with the NDW6 scheme thus far. This simulation examines the impact of cloud microphysics schemes on global cloud-system-resolving simulations in a stepwise fashion. Global cloud-system-resolving models and MMFs, when not coupled with aerosol transport models, have successfully reproduced interannual and intraseasonal variabilities (Khairoutdinov et al. 2008; Satoh et al. 2012). These previous studies have shown that global cloud-system-resolving simulations could improve the climate simulations by explicitly calculating cloud growth. In the next stage, we examine the change in CRF and atmospheric conditions by altering the cloud microphysics scheme in an attempt to determine the physical mechanisms of the process chain that originate from CRF changes. In this study, nucleation processes are calculated by assuming background aerosols. We used a warm cloud nucleation scheme assuming an activated spectrum proposed by Twomey (1959) with a maritime cloud condensation nuclei condition (Seiki and Nakajima 2014). We used the heterogeneous ice nucleation scheme proposed by Philips et al. (2007), which is based on the observational data collected in the free troposphere in the northern midlatitudes (Demott et al. 2003). Hereafter, the simulation using the NDW6 scheme will be referred to as NEW.
3. Observational data To evaluate the simulated hydrometeors, we used the ice water content (IWC) data from the CloudSat level 2B radar-only cloud water content (2B-CWC-RO) product (Austin and Stephens 2001; Austin et al. 2009). In the analysis, we allocated the data into equal-latitude grid boxes of 2.58, and then calculated the zonal mean values and averaged them over June–August (JJA) from 2007 to 2010. Because CloudSat was not launched prior to the simulation period, we used the climatological zonal mean IWCs as a reference. We assumed that the interannual variability in the zonal mean IWCs from the CloudSat 2B-CWC-RO product is smaller than the difference between the observed values and the simulated values. We used the daytime cloud fraction data from the ISCCP D1 product (Rossow and Schiffer, 1999). The
VOLUME 28
cloud fraction is defined by the ratio of cloud pixels to the grid boxes with a size of 2.58 3 2.58; the cloud fraction is categorized by the cloud-top pressure pc into three types: high, middle, and low cloud fraction that are defined by cloudy pixels for which pc , 310 hPa, 310 , pc , 680 hPa, and 680 hPa , pc, respectively. In the simulations, the radiative transfer model includes the ISCCP simulator (Klein and Jakob 1999; Webb et al. 2001) to estimate the equivalent cloud fraction of the ISCCP product. To evaluate the simulated radiative fluxes at the top of the atmosphere and the surface, the radiative fluxes data from the CERES synoptic 18 latitude–longitude gridded edition 3A (SYN1deg_Ed3A) product are used (Wielicki et al. 1996; Kato et al. 2011; Doelling et al. 2013). This product contains integrated observed radiative fluxes from CERES, MODIS, and several geostationary imagers and computed radiative fluxes at the surface based on the Langley Fu–Liou radiative transfer model (Fu and Liou 1992, 1993). The longwave cloud radiative forcing is estimated by calculating the difference between the observed all-sky radiative fluxes and computed clear-sky radiative fluxes. In the comparison, a positive sign indicates a warming effect to Earth’s surface. To compare the precipitation, the precipitation data from the Global Satellite Mapping of Precipitation moving vector with Kalman filter (GSMaP-MVK), version 5, product are used (Ushio et al. 2009); this product was developed by integrating data from the microwave radiometer from the Tropical Rainfall Measuring Mission (TRMM), various microwave radiometers on polar orbital satellites, and infrared radiometers on geostationary satellites using optimization with a Kalman filter method. Because the TRMM orbit is concentrated over the tropics, the analysis region is limited to between 358S and 358N. In the comparison, the observation data are averaged over every grid box with a size of 1.08 3 1.08. We used the Japanese 25-year Reanalysis Project (JRA25) data (Onogi et al. 2007) to reference the atmospheric conditions. For the comparison, the simulated results are linearly interpolated into equal-angle grids and subsequently averaged over each grid box with a size of 2.58 3 2.58. In addition, the vertical coordinates of the simulated results are converted into pressure coordinates with 23 pressure levels in the same manner as the JRA-25 product.
4. Results a. Simulated distributions of hydrometeors and longwave cloud radiative forcing This section briefly summarizes the characteristic differences in the spatial distribution of simulated hydrometeors
15 MARCH 2015
SEIKI ET AL.
2409
FIG. 1. (a) The climatological zonal mean of the observed IWCs (mg m23) averaged from 2007 to 2010, and the zonal mean of the simulated IWCs (mg m23) during JJA in 2004 in the (b) CTL and (c) NEW experiments. Black solid lines are isothermal contours; gray solid lines are isobaric contours. The contours in (a) are obtained from the JRA-25 product.
and LWCRF between the CTL and NEW experiments and the differences between the simulated results and the observed data. Figure 1 shows the vertical distributions of the zonal mean IWCs from the observations and the NEW and CTL experiments. Here, the simulated IWCs include cloud ice, snow, and graupel. The IWC differences between the CTL and NEW experiments are large over the tropics and midlatitudes. In these regions, the IWCs in the NEW experiment are approximately twice as large as the IWCs in the CTL experiment. The zonal mean IWCs in the NEW experiment are closer to the observed values, although interannual variability might be an issue because the observed values are climatological. In the NEW experiment, excessively distributed thin cloud ice in the CTL experiment is clearly removed near the tropopause, and the cloud-top height is similar to the observed height. The numerical setting
FIG. 2. Horizontal distribution and zonal mean of high cloud fractions categorized by the ISCCP D products. (a) The observed daytime visible cloud fraction (%) from the ISCCP D1 product, (b) the simulated cloud fraction by the NSW6 scheme with the ISCCP simulator, and (c) the simulated cloud fraction by the NDW6 scheme with the ISCCP simulator during JJA in 2004.
to switch off the sedimentation of cloud ice in the CTL experiment results in the overestimation of cloud ice near the tropopause. The role of the sedimentation of cloud ice is discussed in detail in section 5c. Figure 2 shows the horizontal distribution of the observed and simulated ISCCP high cloud fraction. The globally averaged values are summarized in Table 1. The globally averaged values of the simulated cloud fractions in the NEW experiment are closer to the observed
2410
JOURNAL OF CLIMATE
TABLE 1. Globally averaged values of the daytime high cloud fraction from ISCCP D1 products and for each experiment.
High cloud fraction
ISCCP
CTL
NEW
0.23
0.27
0.25
values than those in the CTL experiment. The difference in simulated high cloud fraction between the CTL and NEW experiments is large over the tropical ocean and mid- to high-latitude regions. Correspondingly, the IWCs above 200 hPa decrease in the NEW experiment. Figure 3 shows the horizontal distributions of the LWCRF; the globally averaged LWCRFs are summarized in Table 2. The NEW experiment reduces the positive bias over the tropical ocean and Antarctica that is found in the CTL experiment (Figs. 3d,e). The reduction in the LWCRFs is related to the change in the cloud-top height in the NEW experiment, with lower cloud-top heights generally inducing lower LWCRFs.
VOLUME 28
As discussed by Satoh et al. (2010) and Kodama et al. (2012), the terminal velocity of cloud ice and the conversion process of cloud ice to snow strongly affect the amount of cloud ice. The formulation of ice cloud microphysics in the NEW experiment efficiently removes the cloud ice near the tropopause and reduces the positive LWCRFs over such regions. However, these results lead to a negative effect on the globally averaged radiation budget (Table 2) because positive biases over the western Pacific Ocean, Atlantic Ocean, and Maritime Continent and strong negative biases from the eastern Pacific Ocean to South China Sea are cancelled in the CTL experiment (Figs. 3d,f). Similarly, in the NEW experiment, shortwave cloud radiative forcing is reduced over the tropical ocean and becomes similar to the observations; however, the globally averaged shortwave cloud radiative forcing is less accurate than the CTL experiment (not shown). Figure 4 shows the horizontal distributions of precipitation and the corresponding zonal means. In the
FIG. 3. Horizontal distribution and zonal mean of the LWCRF (W m22) during JJA in 2004 based on (c) the CERES synoptic product, and simulations using the (a) NSW6 and (b) NDW6 schemes. The difference between the the observed LWCRF and simulated LWCRF using the (d) NSW6 and (e) NDW6 schemes; (f) the corresponding zonal means are also portrayed.
15 MARCH 2015
SEIKI ET AL.
TABLE 2. Globally averaged differences between simulated LWCRFs and those from the CERES data. The second and third row in each cell show the average values over the Northern and Southern Hemispheres, respectively. DLWCRF (W m22)
Global NH SH
CTL
NEW
NEWFIX
1.31 20.56 3.19
22.01 21.49 22.52
5.16 6.51 3.80
NEW experiment, precipitation over the tropical western Pacific Ocean and northern Indian Ocean is improved by reducing the underestimation in the CTL experiment. Although the intertropical convergence zone (ITCZ) in the CTL experiment has a double peak structure, the ITCZ in the NEW experiment is concentrated in the middle and has a similar structure to the observed ITCZ. These improvements may be linked with the improvements in LWCRF and its effect on atmospheric temperature over the tropics (shown in section 4b) because precipitation is constrained by the energy balances of the atmosphere and the surface (Allen and Ingram 2002). However, precipitation over the tropics is strongly influenced by changes in the horizontal transport of enthalpy flux and LWCRF (Muller and O’Gorman 2011). An examination of the effect of cloud microphysics on the general circulation and changes in the horizontal transport of enthalpy flux is beyond the scope of this study. In the NEW experiment, these improvements in
2411
precipitation with different signs over various regions result in a slight increase in precipitation averaged over the tropics.
b. Reduction in the warm biases near the tropical tropopause The zonal mean vertical profile of atmospheric temperature is strongly modified in the NEW experiment (Fig. 5). In the troposphere, the CTL experiment has a warm bias; the maximum value is approximately 7 K near the lapse-rate tropopause (;100 hPa) over the tropical region and the warm bias is distributed down to the freezing level. Here, the lapse-rate tropopause is defined by the altitude where the temperature lapse rate becomes zero. In the NEW experiment, the warm bias is clearly reduced and the biases are nearly less than 62 K for the entire troposphere. The key factor for this improvement in the NEW experiment is the CRF by cirrus clouds near the tropopause. This subsection examines the origin of the warm bias in the CTL experiment. The radiative heating rate due to the presence of cirrus clouds is shown. Figure 6 shows the time series of the simulated vertical profiles of the heating rate resulting from radiation, phase changes, turbulent mixing, and the difference of the atmospheric temperature relative to the JRA-25 product averaged over the tropics during JJA in 2004. In addition, the mixing ratio of cloud ice is superimposed on those profiles. Here, the tropical region, 108S–108N over all longitudes, is analyzed. Anvil
FIG. 4. Horizontal distribution and zonal mean of precipitation (mm h21) during JJA in 2004 based on (c) the GSMaP_MVK product, and simulations by the (a) NSW6 and (b) NDW6 schemes. The difference between the observed precipitation and simulated precipitation using the (d) NSW6 and the (e) NDW6 schemes; (f) the corresponding zonal means are also portrayed.
2412
JOURNAL OF CLIMATE
VOLUME 28
FIG. 5. The zonal mean vertical profiles of atmospheric temperature averaged during JJA in 2004 from the (a) JRA-25 product, (b) CTL experiment, and (c) NEW experiment. (d),(e) The difference between the simulated atmospheric temperature and the JRA-25 product is shown for each experiment. In (d),(e), the solid lines represent 273-K isotherms and the dashed lines show the lapse rate tropopause.
cirrus clouds expand and cloud ice is concentrated from 300 to 200 hPa in both the CTL and the NEW experiments (Fig. 6). Strong radiative heating occurs at approximately 100 hPa, which is where small amounts of cirrus clouds are distributed. In contrast, the heating rates resulting from phase changes and turbulent mixing are much smaller near the tropopause. The aforementioned features are consistent with the observations and show that the atmospheric heating near the tropopause is dominated by CRF from cirrus clouds (Ackerman et al. 1988; Jensen et al. 1996; Hartmann et al. 2001), which occurs because convective heating and absorption of longwave radiation by carbon dioxide and water vapor is relatively smaller than the absorption of longwave radiation by cirrus clouds above the 200 hPa. In the CTL experiment, the simulated atmospheric temperature continues to increase, particularly around the tropopause, until the simulated atmospheric temperature and radiative fluxes are in approximate balance 30 days into the simulation. In general, the greenhouse effect of cirrus clouds becomes stronger as the cirrus temperature decreases (Liou 2002). Relatively large amounts of cloud ice remains at higher levels in the CTL experiment than in the NEW experiment (Figs. 1 and 6); moreover, weaker radiative cooling is observed in the troposphere in the CTL experiment (Figs. 6a,e). In the middle troposphere, the magnitude of radiative cooling is reduced by an average of 0.2 K day21 during JJA in
2004 compared with the NEW experiment. This reduction results in an overestimation of atmospheric temperature by a few degrees Kelvin during the first 10 days. Therefore, the CTL experiment has a positive atmospheric temperature bias in the troposphere as a result of the cirrus clouds located at approximately 100 hPa, although the total amount of cloud ice is smaller than in the NEW experiment. We discuss the reason why the vertical distribution of cirrus clouds changes because of the alteration of the cloud microphysics scheme in section 5c. Atmospheric temperature biases remain in the NEW experiment, such as those above the tropopause or in the southern midlatitude region. The alteration of the cloud microphysics scheme changes the global atmospheric temperature and does not necessarily improve the regional-scale atmospheric temperature biases. It is beyond the scope of this study to discuss the origin of the regional-scale biases.
5. Discussion a. Role of the cloud optical properties In the NEW experiment, there are two factors that can induce substantial CRF differences based on the CTL experiment: cloud microphysics and cloud optical properties. To isolate the contribution of the two factors to improve atmospheric temperature estimates, an additional
15 MARCH 2015
SEIKI ET AL.
2413
FIG. 6. Time series of the vertical profiles of the heating rate resulting from radiation, phase change, and turbulent mixing (K day21) from (a)–(c) CTL and (e)–(g) NEW experiments and difference between the atmospheric temperature derived from the JRA-25 product and the simulated atmospheric temperature from (d) CTL and (h) NEW experiments (K) averaged over 108S–108N and all longitudes. The contour lines show the mixing ratio of cloud ice (mg kg21).
experiment is performed (called NEWFIX). The experiment is the same as the NEW experiment except that the cloud optical properties are calculated in the same manner as in the CTL experiment; that is, the cloud optical properties are calculated according to Mie theory and the effective radii of liquid and solid hydrometeors are assumed to be 8 and 40 mm, respectively (see section 2a). Figure 7 shows the LWCRFs in the NEWFIX experiment and their differences relative to the CERES product. Table 2 summarizes the globally averaged LWCRFs. The longwave cloud radiative forcings are
obviously overestimated on a global scale. The major contributors to the LWCRF are ice hydrometeors. Compared with the observed IWCs, the IWCs in the NEW experiment are not overestimated and are instead slightly underestimated (Figs. 1a,c). The NEWFIX experiment has been suggested to overestimate the LWCRF by assuming effective radii of solid hydrometeors as 40 mm in the calculation of the cloud optical properties. Recent progress in satellite analyses has revealed global vertical distributions of the effective radii of solid hydrometeors (Okamoto et al. 2010; Sato and Okamoto 2011). According to the observed profiles, the
2414
JOURNAL OF CLIMATE
VOLUME 28
FIG. 7. As in Fig. 3a,b, but for the NEWFIX experiment.
effective radii below 10 km (;300 hPa) exceed 100 mm over the tropics. Thus, the assumed effective radii in the CTL and NEWFIX experiments are inconsistent with the relatively large observed effective radii, which originate from the presence of the precipitating solid hydrometeors. Figure 8 shows the zonal mean vertical profile of atmospheric temperature in the NEWFIX experiment and its difference relative to the JRA-25 product. The general features of the atmospheric temperature bias are quite similar to the NEW experiment, although the radiative budget at the top of the atmosphere increasingly differs from the observations in the NEWFIX experiment. Comparing the CRFs between the NEW and NEWFIX experiments, the CRF change that results from the assumed cloud optical properties in the radiative transfer calculation is found to be 7 W m22 for the LWCRF. The extent of the change is greater than the change in the CRF that results from the alteration of cloud microphysics scheme (see Table 2). Careful attention is necessary to investigate process chains involving CRF using such simple cloud microphysics schemes (e.g., the NSW6 scheme) that assume fixed effective radii for calculating the cloud optical properties because tuning globally
averaged CRFs (e.g., Kodama et al. 2012) could conceal cloud microphysics biases.
b. Persistence of cirrus clouds Here, we discuss the contribution of cloud ice to the radiative heating rate near the tropopause over the tropics. The domain extending over 108S–108N and all longitudes from 200 to 100 hPa is analyzed. Figure 9a shows the cloud fraction sorted by the cloud ice mixing ratio CF(qi), which is defined by the ratio of the gridscale air mass for each value of qi against the air mass of the entire domain as follows:
CF(qi ) 5
å
rV
å
rV
qi ,q#qi 1Dq qmin ,q#qmax
,
where V is the volume of a grid box, qmin 5 0.1 mg kg21, qmax 5 100 mg kg21, and qi is discretized into 30 bins within a range from qmin to qmax on a logarithmic scale. In the CTL experiment, there is a turning point at qi 5 5 mg kg21. In contrast, a moderate distribution of the cloud fraction is produced in the NEW experiment, which has a mode qi of approximately 1.4 mg kg21,
FIG. 8. As in Fig. 5b,d, but for the NEWFIX experiments.
15 MARCH 2015
SEIKI ET AL.
2415
FIG. 9. (a) The cloud fraction sorted by qi and (b) domain-averaged total radiative heating rate sorted by qi in each experiment.
which is smaller than that produced in the CTL experiment. Figure 9b shows the domain-averaged total radiative heating rate sorted by the cloud ice mixing ratio QRADave(qi), which is defined as follows:
å
QRADave (qi ) 5
qi ,q#qi 1Dq
å
rVQRAD
qmin ,q#qmax
rV
,
where QRAD is the radiative heating rate in a grid box. As shown in Fig. 9a, there is a turning point at qi 5 qicrt in the CTL experiment. Because the radiative heating rate of cirrus clouds increases as qi increases, the large qi mode in the cloud fraction in the CTL experiment leads to a large radiative heating rate near the tropopause. In the NEW experiment, the cirrus cloud fraction for qi exceeding 1 mg kg21 is efficiently removed. Therefore, the radiative heating rate near the tropopause decreases. As mentioned in section 2, the conversion from cloud ice to snow is initiated at an empirical critical value of the cloud ice mixing ratio of qicrt 5 5 mg kg21, assuming a time scale for the autoconversion (t auto) in the NSW6 scheme (Kodama et al. 2012). The artificial turning points observed in the cloud fraction and radiative heating rate shown in Figs. 9a and 9b correspond to the empirical value of qicrt. In contrast, in the NEW experiment, the autoconversion from cloud ice to snow is resolved using an approximated stochastic collection equation instead of the simple linear dumping, which is used in the NSW6 scheme (see the appendix). This theoretical formulation of the conversion rate provides for the continuous growth of ice particles by aggregation depending on their size and number concentration. As a result, the radiative heating rate near the tropopause is reduced; moreover, the zonal mean profiles of atmospheric temperature are improved in the NEW experiment. We then speculate as to why the ice particles are persistent near the tropopause in the CTL experiment. Large values of qi might be removed from the tropopause by gravitational sedimentation or sublimation, although the autoconversion from the cloud ice to snow is not sufficiently efficient for dissipation. Jensen et al.
(1996) suggested that the persistence of cirrus clouds near the tropical tropopause is strongly influenced by sublimation and the lofting of cloudy parcels by buoyancy that originates from radiative heating. The former works to decreases the occurrence of cirrus clouds, whereas the latter works to enlarge cirrus clouds. In the CTL experiment, large amounts of thin cirrus clouds near the tropopause can be found, which means that the lofting effect by radiative heating works efficiently to maintain persistent cirrus clouds. The lofting effect can be estimated according to the sum of the updraft and sedimentation. In the CTL experiment, the sedimentation of cloud ice is artificially switched off to adjust for the simulated CRF at the top of the atmosphere to the observed value (Kodama et al. 2012). In the NEW experiment, the average qi fallout is approximately 720 m h21 (20 cm s21) at qi 5 5 mg kg21 (Fig. 10). In the context of Jensen’s suggestion, the lack of sedimentation overestimates the lofting effect and makes it more persistent in the CTL experiment, particularly over the tropics. The warm bias in the troposphere in the CTL experiment is suggested to originate from two factors in the NSW6 scheme: the empirical formulation of the autoconversion from cloud ice to snow and the sedimentation of cloud ice. Kodama et al. (2012) reported that smaller values of qicrt and the implementation of cloud ice sedimentation reduce the cloud fraction of thin cirrus clouds and lower the cirrus cloud-top height; therefore, the CRF by ice clouds is underestimated. A sensitivity experiment that is the same as the CTL experiment except that the sedimentation of cloud ice is calculated and qicrt is set to 0 [the qicrt0–4water experiment in Kodama et al. (2012)] has shown smaller biases in atmospheric temperature during the first 10 days of the simulation (not shown). However, the high cloud fraction is reduced by half and CRFs are underestimated (approximately 13.1 W m22 for the LWCRF). A similar sensitivity of the terminal velocity of cloud ice to the CRF was obtained by Heymsfield and Iaquinta (2000). Rougier et al. (2009) showed that the choice of small values for the terminal velocity of cloud ice increases the
2416
JOURNAL OF CLIMATE
FIG. 10. The domain-averaged mass-weighted terminal velocity sorted by qi in the NEW experiment.
climate sensitivity in the Hadley Centre Slab Climate Model version 3. Thus, the radiation budget and persistence of cirrus clouds are very sensitive to the balance between the autoconversion and sedimentation processes. Here, the balance between the two factors in the NEW experiment is not necessarily accurate because the two factors are not directly validated by the observations, although atmospheric temperature is well reproduced. We expect that the new earth observation satellite EarthCARE, which has been developed by the European Space Agency, Japan Aerospace Exploration Agency, and National Institute of Information and Communications Technology and is planned to be launched in 2015, will provide valuable data to constrain the terminal velocity of ice particles using Doppler radar.
6. Conclusions We examined the impact of the alteration of the cloud microphysics scheme on the distribution of cloud fraction, LWCRF, and atmospheric temperature in global cloud-system-resolving simulations over three months. The NDW6 scheme was relatively accurate at reproducing the vertical distribution of IWC and cirrus cloud-top heights. Improvements were also shown in the horizontal distribution of high clouds. However, the globally averaged LWCRF exhibit additional bias despite decreases in the high cloud fraction biases. The improvement in the horizontal distribution of precipitation is similar to that of LWCRF. To reduce the remaining biases, further investigations into other cloud microphysical processes and related physical processes are necessary in the future. In the CTL experiment, atmospheric temperature has warm biases of up to approximately 7 K that are primarily located near the tropopause. A change in the cloud optical properties works to slightly reduce the warm biases; however, it does not dominantly influence atmospheric temperature despite the substantial change in the radiation budget. The warm biases can be greatly reduced using the NDW6 scheme through the cloud microphysical response and associated cirrus LWCRF. The amount of
VOLUME 28
high clouds quickly responds to the alteration of the cloud microphysics scheme within the first few days of the simulation. In the CTL experiment, the warm trend continues until the atmospheric temperature balances at a warmer state after 30 days. The overestimation of the high cloud fraction causes a greenhouse effect that induces a warm bias in the troposphere. The efficient removal of cloud ice by aggregation and sedimentation processes are key processes for reducing the bias. We suggest the use of theoretical formulations as used in the NDW6 scheme to represent the detailed dependences of growth rates on microphysical parameters (see the appendix). Acknowledgments. The authors wish to thank the NICAM development members for their helpful comments and discussions. We thank the three anonymous reviewers for their comments that helped to improve the manuscript. The satellite datasets from A-Train were obtained from the NASA Langley Research Center Atmospheric Science Data Center. The ISCCP D1 data were obtained from the International Satellite Cloud Climatology Project web site (http://isccp.giss.nasa.gov) maintained by the ISCCP research group at the NASA Goddard Institute for Space Studies, New York, NY, in January 2005. ‘‘The reanalysis dataset of Global Rainfall Map (GSMaP_MVK) by JAXA Global Rainfall Watch’’ was produced and distributed by the Earth Observation Research Center, Japan Aerospace Exploration Agency. The numerical experiments were performed using The Earth Simulator at the Japan Agency for Marine-Earth Science and Technology. The present study was supported by ‘‘Prediction and Diagnosis of Imminent Global Climate Change’’ in the Program for Risk Information on Climate Change (SOUSEI Program), funded by the Ministry of Education, Culture, Sports, Science, and Technology (MEXT).
APPENDIX Autoconversion from Cloud Ice to Snow In the NSW6 scheme, the conversion from cloud ice to snow is parameterized as follows: ›rqi r(qi 2 qicrt ) 5 2Ec (T) , t auto ›t
(A1)
where Ec is the sticking efficiency as a function of atmospheric temperature, qicrt is a critical value to initiate the conversion process, and t auto is the relaxation time scale of the conversion process. Kodama et al. (2012) assumed that the values of qicrt and t auto are global constants.
15 MARCH 2015
SEIKI ET AL.
In the NDW6 scheme, based on Seifert and Beheng (2006), the stochastic collection equation is solved by assuming a binary collision to derive the conversion rate from cloud ice to snow via aggregation of cloud ice as follows: ð‘ ð‘ ›rqi 5 2Ec (T) fi (x)fi (y)xAi (x, y)jdy i (x, y)j dx dy, ›t 0 0 (A2) where fi(x) is the mass distribution function of cloud ice with a particle mass of x, Ai is the collisional cross section, and jdyij is the relative terminal velocity between two colliding particles. In the NDW6 scheme, this equation is solved with several approximations. Seifert and Beheng (2006) should be referred to for the detailed derivation. Here, we attempt to show a simplified dependence of the growth rate on the bulk cloud parameters with further approximations. Assuming a monodisperse mass distribution function for f(x), the mean projected area of cloud ice Ai represents Ai, and the mean terminal velocity of cloud ice yi represents jdy ij, Eq. (A2) can be solved as follows: ›rqi ; 2Ec (T)rqi Ni Ai y i . ›t
(A3)
Furthermore, assuming that cloud ice consists of spherical particles that are sufficiently small to satisfy Stokes’ law for evaluating their terminal velocities, the dependence of the growth rate is approximated as follows: ›rqi 4 } 2Ec (T)rqi Ni Di , ›t
(A4)
where Di is the mean mass diameter. Comparing Eq. (A4) with Eq. (A1), the critical value speculated to initiate the conversion process (qicrt) results from the strong dependence of the conversion rate on the mean mass diameter. In the theoretical formulation, the relaxation time scale and critical value are not global constants; instead, these values are dependent on the mixing ratio, number concentration, and mean mass diameter. REFERENCES Abel, S. J., and I. A. Boutle, 2012: An improved representation of the raindrop size distribution for single-moment microphysics schemes. Quart. J. Roy. Meteor. Soc., 138, 2151–2162, doi:10.1002/ qj.1949. Ackerman, T. P., K.-N. Liou, F. P. J. Valero, and L. Pfister, 1988: Heating rates in tropical anvils. J. Atmos. Sci., 45, 1606–1623, doi:10.1175/1520-0469(1988)045,1606:HRITA.2.0.CO;2. Allen, M. R., and W. J. Ingram, 2002: Constraints on future changes in climate and the hydrological cycle. Nature, 419, 224–232, doi:10.1038/nature01092.
2417
Austin, R. T., and G. L. Stephens, 2001: Retrieval of stratus cloud microphysical parameters using millimeter-wave radar and visible optical depth in preparation for CloudSat: 1. Algorithm formulation. J. Geophys. Res., 106, 28 233–28 242, doi:10.1029/ 2000JD000293. ——, A. J. Heymsfield, and G. L. Stephens, 2009: Retrieval of ice cloud microphysical parameters using the CloudSat millimeter-wave radar and temperature. J. Geophys. Res., 114, D00A23, doi:10.1029/2008JD010049. Berry, E. X., and R. L. Reinhardt, 1974: An analysis of cloud drop growth by collection. Part II: Single initial distributions. J. Atmos. Sci., 31, 1825–1831, doi:10.1175/1520-0469(1974)031,1825: AAOCDG.2.0.CO;2. Cess, R. D., and Coauthors, 1989: Interpretation of cloud–climate feedback as produced by 14 atmospheric general circulation models. Science, 245, 513–516, doi:10.1126/science.245.4917.513. Demott, P. J., D. J. Cziczo, A. J. Prenni, D. M. Murphy, S. M. Kreidenweis, D. S. Thompson, R. Borys, and D. C. Rogers, 2003: Measurements of the concentration and composition of nuclei for cirrus formation. Proc. Natl. Acad. Sci. USA, 100, 14 655–14 660, doi:10.1073/pnas.2532677100. Doelling, D. R., and Coauthors, 2013: Geostationary enhanced temporal interpolation for CERES flux products. J. Atmos. Oceanic Technol., 30, 1072–1090, doi:10.1175/JTECH-D-12-00136.1. Field, P. R., R. J. Hogan, P. R. A. Brown, A. J. Illingworth, T. W. Choularton, and R. J. Cotton, 2005: Parameterization of iceparticle size distributions for mid-latitude stratiform cloud. Quart. J. Roy. Meteor. Soc., 131, 1997–2017, doi:10.1256/ qj.04.134. ——, A. J. Heymsfield, and A. Bansemer, 2007: Snow size distribution parameterization for midlatitude and tropical ice clouds. J. Atmos. Sci., 64, 4346–4365, doi:10.1175/ 2007JAS2344.1. Fu, Q., 1996: An accurate parameterization of the solar radiative properties of cirrus clouds for climate models. J. Climate, 9, 2058–2082, doi:10.1175/1520-0442(1996)009,2058:AAPOTS.2.0.CO;2. ——, and K. N. Liou, 1992: On the correlated k-distribution method for radiative transfer in nonhomogeneous atmospheres. J. Atmos. Sci., 49, 2139–2156, doi:10.1175/1520-0469(1992)049,2139: OTCDMF.2.0.CO;2. ——, and L. N. Liou, 1993: Parameterization of the radiative properties of cirrus clouds. J. Atmos. Sci., 50, 2008–2025, doi:10.1175/ 1520-0469(1993)050,2008:POTRPO.2.0.CO;2. ——, P. Yang, and W. B. Sun, 1998: An accurate parameterization of the infrared radiative properties of cirrus clouds for climate models. J. Climate, 11, 2223–2237, doi:10.1175/1520-0442(1998)011,2223: AAPOTI.2.0.CO;2. Gettelman, A., H. Morrison, and S. J. Ghan, 2008: A new two-moment bulk stratiform cloud microphysics scheme in the Community Atmospheric Model, version 3 (CAM3). Part II: Single-column and global results. J. Climate, 21, 3660–3679, doi:10.1175/ 2008JCLI2116.1. Grabowski, G. G., 1998: Toward cloud resolving modeling of largescale tropical circulation: A simple cloud microphysics parameterization. J. Atmos. Sci., 55, 3283–3298, doi:10.1175/ 1520-0469(1998)055,3283:TCRMOL.2.0.CO;2. ——, 2001: Coupling cloud processes with the large-scale dynamics using the cloud-resolving convection parameterization (CRCP). J. Atmos. Sci., 58, 978–997, doi:10.1175/1520-0469(2001)058,0978: CCPWTL.2.0.CO;2. Hartmann, D. L., J. R. Holton, and Q. Fu, 2001: The heat balance of the tropical tropopause, cirrus, and stratospheric dehydration. Geophys. Res. Lett., 28, 1969–1972, doi:10.1029/2000GL012833.
2418
JOURNAL OF CLIMATE
Hashino, T., M. Satoh, Y. Hagihara, T. Kubota, T. Matsui, T. Nasuno, and H. Okamoto, 2013: Evaluating cloud microphysics from NICAM against CloudSat and CALIPSO. J. Geophys. Res. Atmos., 118, 7273–7292, doi:10.1002/jgrd.50564. Heymsfield, A. J., and J. Iaquinta, 2000: Cirrus crystal terminal velocities. J. Atmos. Sci., 57, 916–938, doi:10.1175/ 1520-0469(2000)057,0916:CCTV.2.0.CO;2. Iga, S., H. Tomita, Y. Tsushima, and M. Satoh, 2007: Climatology of a nonhydrostatic global model with explicit cloud processes. Geophys. Res. Lett., 34, L22814, doi:10.1029/2007GL031048. Ikawa, M., H. Mizuno, T. Matsuo, M. Murakami, Y. Yamada, and K. Saito, 1991: Numerical modeling of the convective snow cloud over the Sea of Japan—Precipitation mechanism and sensitivity to ice crystal concentration rates. J. Meteor. Soc. Japan, 69, 641–667. Jensen, E. J., O. B. Toon, H. B. Selkirk, J. D. Spinhirne, and M. R. Schoeberl, 1996: On the formation and persistence of subvisible cirrus clouds near the tropical tropopause. J. Geophys. Res., 101, 21 361–21 375, doi:10.1029/95JD03575. Kato, S., and Coauthors, 2011: Improvements of top-of-atmosphere and surface irradiance computations with CALIPSO-, CloudSat-, and MODIS-derived cloud and aerosol properties. J. Geophys. Res., 116, D19209, doi:10.1029/2011JD016050. Khairoutdinov, M. F., and D. A. Randall, 2003: Cloud resolving modeling of the ARM summer 1997 IOP: Model formulation, results, uncertainties, and sensitivities. J. Atmos. Sci., 60, 607–625, doi:10.1175/1520-0469(2003)060,0607:CRMOTA.2.0.CO;2. ——, C. DeMott, and D. A. Randall, 2008: Evaluation of the simulated interannual and subseasonal variability in an AMIP-style simulation using the CSU Multiscale Modeling Framework. J. Climate, 21, 413–431, doi:10.1175/2007JCLI1630.1. Klein, S., and C. Jakob, 1999: Validation and sensitivities of frontal clouds simulated by the ECMWF model. Mon. Wea. Rev., 127, 2514–2531, doi:10.1175/1520-0493(1999)127,2514: VASOFC.2.0.CO;2. Kodama, C., A. T. Noda, and M. Satoh, 2012: An assessment of the cloud signals simulated by NICAM using ISCCP, CALIPSO, and CloudSat satellite simulators. J. Geophys. Res., 117, D12210, doi:10.1029/2011JD017317. Lang, S. E., W.-K. Tao, X. Zeng, and Y. Li, 2011: Reducing the biases in simulated radar reflectivities from a bulk microphysics scheme: Tropical convective systems. J. Atmos. Sci., 68, 2306–2320, doi:10.1175/JAS-D-10-05000.1. ——, ——, J. Chern, D. Wu, and X. Li, 2014: Benefits of a fourth ice class in the simulated radar reflectivities of convective systems using a bulk microphysics scheme. J. Atmos. Sci., 71, 3583– 3612, doi:10.1175/JAS-D-13-0330.1. Lin, Y.-L., R. D. Farley, and H. D. Orville, 1983: Bulk parameterization of the snow field in a cloud model. J. Climate Appl. Meteor., 22, 1065– 1092, doi:10.1175/1520-0450(1983)022,1065:BPOTSF.2.0.CO;2. Liou, K.-N., 2002: An Introduction to Atmospheric Radiation. Academic Press, 583 pp. Louis, J. F., 1979: A parametric model of vertical eddy fluxes in the atmosphere. Bound.-Layer Meteor., 17, 187–202, doi:10.1007/ BF00117978. Mansell, E. R., 2010: On sedimentation and advection in multimoment bulk microphysics. J. Atmos. Sci., 67, 3084–3094, doi:10.1175/ 2010JAS3341.1. Martin, G. M., S. F. Milton, C. A. Senior, M. E. Brooks, S. Ineson, T. Reichler, and J. Kim, 2010: Analysis and reduction of systematic errors through a seamless approach to modeling weather and climate. J. Climate, 23, 5933–5957, doi:10.1175/ 2010JCLI3541.1.
VOLUME 28
Matsui, T., X. Zeng, W.-K. Tao, H. Masunaga, W. S. Olson, and S. Lang, 2009: Evaluation of long-term cloud-resolving model simulations using satellite radiance observations and multifrequency satellite simulators. J. Atmos. Oceanic Technol., 26, 1261–1274, doi:10.1175/2008JTECHA1168.1. Milbrandt, J. A., and M. K. Yau, 2005: A multimoment bulk microphysics parameterization. Part I: Analysis of the role of the spectral shape parameter. J. Atmos. Sci., 62, 3051–3064, doi:10.1175/ JAS3534.1. ——, and R. McTaggart-Cowan, 2010: Sedimentation-induced errors in bulk microphysics schemes. J. Atmos. Sci., 67, 3931– 3948, doi:10.1175/2010JAS3541.1. Mitchell, D. L., 1996: Use of mass- and area-dimensional power laws for determining precipitation particle terminal velocities. J. Atmos. Sci., 53, 1710–1723, doi:10.1175/1520-0469(1996)053,1710: UOMAAD.2.0.CO;2. Morrison, H., and A. Gettelman, 2008: A new two-moment bulk stratiform cloud microphysics scheme in the Community Atmosphere Model, version 3 (CAM3). Part I: Description and numerical tests. J. Climate, 21, 3642–3659, doi:10.1175/ 2008JCLI2105.1. Muller, C. J., and P. A. O’Gorman, 2011: An energetic perspective on the regional response of precipitation to climate change. Nat. Climate Change, 1, 266–271, doi:10.1038/ nclimate1169. Nakajima, T., M. Tsukamoto, Y. Tsushima, A. Numaguchi, and T. Kimura, 2000: Modeling of the radiative process in an atmospheric general circulation model. Appl. Opt., 39, 4869– 4878, doi:10.1364/AO.39.004869. Nakanishi, M., and H. Niino, 2004: An improved Mellor-Yamada level-3 model with condensation physics: Its design and verification. Bound.-Layer Meteor., 112, 1–31, doi:10.1023/B: BOUN.0000020164.04146.98. Noda, A. T., K. Oouchi, M. Satoh, H. Tomita, S. Iga, and Y. Tsushima, 2010: Importance of the subgrid-scale turbulent moist process: Cloud distribution in global cloud-resolving simulations. Atmos. Res., 96, 208–217, doi:10.1016/j.atmosres.2009.05.007. Okamoto, H., K. Sato, and Y. Hagihara, 2010: Global analysis of ice microphysics from CloudSat and CALIPSO: Incorporation of specular reflection in lidar signals. J. Geophys. Res., 115, D22209, doi:10.1029/2009JD013383. Onogi, K., and Coauthors, 2007: The JRA-25 Reanalysis. J. Meteor. Soc. Japan, 85, 369–432, doi:10.2151/jmsj.85.369. Philips, V. T. J., L. J. Donner, and S. T. Garner, 2007: Nucleation processes in deep convection simulated by a cloud-system-resolving model with double-moment bulk microphysics. J. Atmos. Sci., 64, 738–761, doi:10.1175/JAS3869.1. Phillips, T. J., and Coauthors, 2004: Evaluating parameterizations in general circulation models: Climate simulation meets weather prediction. Bull. Amer. Meteor. Soc., 85, 1903–1915, doi:10.1175/BAMS-85-12-1903. Roh, W., and M. Satoh, 2014: Evaluation of precipitating hydrometeor parameterizations in a single-moment bulk microphysics scheme for deep convective systems over the tropical central Pacific. J. Atmos. Sci., 71, 2654–2673, doi:10.1175/ JAS-D-13-0252.1. Rossow, B. R., and R. A. Schiffer, 1999: Advances in understanding clouds from ISCCP. Bull. Amer. Meteor. Soc., 80, 2261–2287, doi:10.1175/1520-0477(1999)080,2261:AIUCFI.2.0.CO;2. Rougier, J., D. M. H. Sexton, J. M. Murphy, and D. Stainforth, 2009: Analyzing the climate sensitivity of the HadSM3 climate model using ensembles from different but related experiments. J. Climate, 22, 3540–3557, doi:10.1175/2008JCLI2533.1.
15 MARCH 2015
SEIKI ET AL.
Rutledge, S. A., and P. V. Hobbs, 1983: The mesoscale and microscale structure and organization of clouds and precipitation in midlatitude cyclones. VIII: A model for the ‘‘seeder-feeder’’ process in warm frontal rainbands. J. Atmos. Sci., 40, 1185–1206, doi:10.1175/1520-0469(1983)040,1185: TMAMSA.2.0.CO;2. Salzmann, M., Y. Ming, J.-C. Golaz, P. A. Ginoux, H. Morrison, A. Gettelman, M. Krämer, and L. J. Donner, 2010: Twomoment bulk stratiform cloud microphysics in the GFDL AM3 GCM: Description, evaluation, and sensitivity tests. Atmos. Chem. Phys., 10, 8037–8064, doi:10.5194/acp-10-8037-2010. Sato, K., and H. Okamoto, 2011: Refinement of global ice microphysics using spaceborne active sensors. J. Geophys. Res., 116, D20202, doi:10.1029/2011JD015885. Satoh, M., T. Matsuno, H. Tomita, H. Miura, T. Nasuno, and S. Iga, 2008: Nonhydrostatic Icosahedral Atmospheric Model (NICAM) for global cloud resolving simulations. J. Comput. Phys., 227, 3486–3514, doi:10.1016/j.jcp.2007.02.006. ——, T. Inoue, and H. Miura, 2010: Evaluation of cloud properties of global and local cloud system resolving model using CALIPSO and CloudSat simulators. J. Geophys. Res., 115, D00H14, doi:10.1029/2009JD012247. ——, and Coauthors, 2012: The Intra-Seasonal Oscillation and its control of tropical cyclones simulated by high-resolution global atmospheric models. Climate Dyn., 39, 2185–2206, doi:10.1007/s00382-011-1235-6. ——, and Coauthors, 2014: The Non-hydrostatic Icosahedral Atmospheric Model: Description and development. Prog. Earth Planet. Sci., 1 (18), doi:10.1186/s40645-014-0018-1. Seifert, A., 2008: On the parameterization of evaporation of raindrops as simulated by a one-dimensional rainshaft model. J. Atmos. Sci., 65, 3608–3619, doi:10.1175/2008JAS2586.1. ——, and K. D. Beheng, 2001: A double-moment parameterization for simulating autoconversion, accretion, and selfcollection. Atmos. Res., 59-60, 265–281, doi:10.1016/S0169-8095(01)00126-0. ——, and ——, 2006: A two-moment cloud microphysics parameterization for mixed-phase clouds. Part I: Model description. Meteor. Atmos. Phys., 92, 45–66, doi:10.1007/s00703-005-0112-4. ——, A. Khain, A. Pokrovsky, and K. D. Beheng, 2006: A comparison of spectral bin and two-moment bulk mixed-phase cloud microphysics. Atmos. Res., 80, 46–66, doi:10.1016/ j.atmosres.2005.06.009. Seiki, T., and T. Nakajima, 2014: Aerosol effects of the condensation process on a convective cloud simulation. J. Atmos. Sci., 71, 833–853, doi:10.1175/JAS-D-12-0195.1. ——, M. Satoh, H. Tomita, and T. Nakajima, 2014: Simultaneous evaluation of ice cloud microphysics and nonsphericity of the cloud optical properties using hydrometeor video sonde and radiometer sonde in situ observations. J. Geophys. Res. Atmos., 119, 6681–6701, doi:10.1002/2013JD021086. Sekiguchi, M., and T. Nakajima, 2008: A k-distribution-based radiation code and its computational optimization for an atmospheric general circulation model. J. Quant. Spectrosc. Radiat. Transfer, 109, 2779–2793, doi:10.1016/j.jqsrt.2008.07.013.
2419
Takata, K., S. Emori, and T. Watanabe, 2003: Development of the minimal advanced treatments of surface interaction and runoff. Global Planet. Change, 38, 209–222, doi:10.1016/ S0921-8181(03)00030-4. Tao, W.-K., and Coauthors, 2009: A multiscale modeling system: Developments, applications, and critical issues. Bull. Amer. Meteor. Soc., 90, 515–534, doi:10.1175/2008BAMS2542.1. ——, and Coauthors, 2014: The Goddard Cumulus Ensemble model (GCE): Improvements and applications for studying precipitation processes. Atmos. Res., 143, 392–424, doi:10.1016/ j.atmosres.2014.03.005. Thompson, G., P. R. Field, R. M. Rasmussen, and W. D. Hall, 2008: Explicit forecasts of winter precipitation using an improved bulk microphysics scheme. Part II: Implementation of a new snow parameterization. Mon. Wea. Rev., 136, 5095–5115, doi:10.1175/ 2008MWR2387.1. Tomita, H., 2008: New microphysical schemes with five and six categories by diagnostic generation of cloud ice. J. Meteor. Soc. Japan, 86A, 121–142, doi:10.2151/jmsj.86A.121. ——, and M. Satoh, 2004: A new dynamical framework of nonhydrostatic global model using the icosahedral grid. Fluid Dyn. Res., 34, 357–400, doi:10.1016/j.fluiddyn.2004.03.003. Twomey, S., 1959: The nuclei of natural cloud formation. Part II: The supersaturation in natural clouds and the variation of cloud droplet concentration. Pure Appl. Geophys., 43, 243– 249, doi:10.1007/BF01993560. Ushio, T., and Coauthors, 2009: A Kalman filter approach to the Global Satellite Mapping of Precipitation (GSMaP) from combined passive microwave and infrared radiometric data. J. Meteor. Soc. Japan, 87A, 137–151, doi:10.2151/jmsj.87A.137. Wacker, W., and A. Seifert, 2001: Evolution of rain water profiles resulting from pure sedimentation: Spectral vs. parameterized description. Atmos. Res., 58, 19–39, doi:10.1016/ S0169-8095(01)00081-3. Wang, M., and Coauthors, 2011a: The multi-scale aerosol-climate model PNNL-MMF: Model description and evaluation. Geosci. Model Dev., 4, 137–168, doi:10.5194/gmd-4-137-2011. ——, S. Ghan, M. Ovchinnikov, X. Liu, R. Easter, E. Kassianov, Y. Qian, and H. Morrison, 2011b: Aerosol indirect effects in a multi-scale aerosol-climate model PNNL-MMF. Atmos. Chem. Phys., 11, 5431–5455, doi:10.5194/acp-11-5431-2011. Webb, M., C. Senior, S. Bony, and J. J. Morcrette, 2001: Combining ERBE and ISCCP data to assess clouds in the Hadley Centre, ECMWF and LMD atmosphere climate models. Climate Dyn., 17, 905–922, doi:10.1007/s003820100157. Wielicki, B. A., B. R. Barkstrom, E. F. Harrison, R. B. Lee, G. L. Smith, and J. E. Cooper, 1996: Clouds and the Earth’s Radiant Energy System (CERES): An Earth observing system experiment. Bull. Amer. Meteor. Soc., 77, 853–868, doi:10.1175/ 1520-0477(1996)077,0853:CATERE.2.0.CO;2. Williams, K. D., and Coauthors, 2013: The Transpose-AMIP II experiment and its application to the understanding of Southern Ocean cloud biases in climate models. J. Climate, 26, 3258–3274, doi:10.1175/JCLI-D-12-00429.1.
