The GOCART–Thompson microphysics scheme coupling the GOCART aerosol model and the aerosol-aware Thompson–Eidhammer microphysics scheme has been implemented in the WRF-Chem to quantify and evaluate the effect of dust on the ice nucleation process in the atmosphere by serving as ice nuclei (IN). The performance of the GOCART–Thompson microphysics scheme in simulating the effect of dust in atmospheric ice nucleation is then evaluated over East Asia during spring, a typical dust-intensive season, in 2012. Based upon the dust emission reasonably reproduced by WRF-Chem, the effect of dust on atmospheric cloud ice water content is well reproduced. With abundant dust particles serving as IN, the simulated ice water mixing ratio and ice crystal number concentration increases by 15 and 7 % on average over the dust source region and downwind areas during the investigated period. The comparison with the ice water path from satellite observations demonstrated that the simulation of the cloud ice profile is substantially improved by considering the indirect effect of dust particles in the simulations. Additional sensitivity experiments are carried out to optimize the parameters in the ice nucleation parameterization in the GOCART–Thompson microphysics scheme. Results suggest that lowering the threshold relative humidity with respect to ice to 100 % for the ice nucleation parameterization leads to further improvement in cloud ice simulation.
Dust aerosol is the second largest contributor to the global aerosol burden (Textor et al., 2006), and it is estimated to contribute around 20 % to annual global aerosol emissions (Tomasi et al., 2017). The Intergovernmental Panel on Climate Change (IPCC) has recognized dust as a major component of atmospheric aerosols, which are an “essential climate variable.” East Asia is a main contributor to the Earth's dust emissions. It has been reported in previous studies that East Asian dust contributes 25–50 % of global emissions, depending on the climate of the particular year (Ginoux et al., 2001).
Dust in the atmosphere alters the Earth's weather and climate in certain ways. By reflecting, absorbing, and scattering the incoming solar radiation, dust can cause a warming effect within the atmosphere and a cooling effect at the surface layer (Lacis, 1995), which is the direct effect of dust. The semi-direct effect of dust is related to the absorption of shortwave and longwave radiation by dust aerosol within clouds, leading to a warming of the surrounding environment and causing a shrinking of cloud and a lower cloud albedo, thus modifying the radiation budget (Perlwitz and Miller, 2010; Hansen et al., 1997). The dust–cloud interaction is also referred to as the indirect effect of dust. Dust particles are recognized as effective IN and play an important role in the ice nucleation process in the atmosphere, directly affecting the dynamics in ice and mixed-phase clouds, such as the formation and development of clouds and precipitation (Koehler et al., 2010; Twohy et al., 2009).
To date, many studies have been conducted to evaluate the direct radiative effect of dust aerosol using radiation schemes implemented in numerical models all over the world (Mallet et al., 2009; Nabat et al., 2015a; Ge et al., 2010; Hartmann et al., 2013; Huang et al., 2009; Bi et al., 2013; Y. Liu et al., 2011; J. Liu et al., 2011; Chen et al., 2017). Recently, the semi-direct effect of dust has been investigated in a few studies over different regions by applying various global and regional models (Tesfaye et al., 2015; Nabat et al., 2015b; Seigel et al., 2013). Unfortunately, due to the poor understanding of the dust–cloud interactions in microphysics processes, quantifying the microphysical effect of dust remains a difficult problem. Various ice nucleation parameterizations have been implemented into global models to estimate the importance of dust in atmospheric ice nucleation (Lohmann and Diehl, 2006; Karydis et al., 2011; Hoose et al., 2008; Zhang et al., 2014). However, most regional models are not capable of estimating the indirect effect of dust, and very rarely has work been done to assess the indirect effects of dust on the weather system, especially over East Asia, which is a major contributor to global dust emissions. Currently, only a few microphysics schemes considering the aerosol–cloud interaction are implemented in regional models. In most of these microphysics schemes only the cloud condensation nuclei (CCN) served by aerosols are considered (Perlwitz and Miller, 2010; Solomos et al., 2011; Miller et al., 2004), while IN are not treated or represented by a prescribed IN distribution (Chapman et al., 2009; Baró et al., 2015), and the production of ice crystals is simplified by a function of temperature or ice saturation. In reality, however, the number of ice crystals that can form in the atmosphere is highly dependent on the number of particles that can act as IN, and dust is the most abundant aerosol that can effectively serve as IN and affect the formation and development of mixed-phase and ice clouds in the atmosphere. This effect should not be neglected in numerical models, especially in the simulations over arid regions during strong wind events (DeMott et al., 2003, 2015; Koehler et al., 2010; Lohmann and Diehl, 2006; Atkinson et al., 2013).
In 2014, the aerosol-aware Thompson–Eidhammer microphysics scheme, which takes into account the aerosols serving as CCN and IN, has been implemented into the Weather Research and Forecast (WRF) model and also the Weather Research and Forecast model coupled with Chemistry (WRF-Chem), enabling the model to explicitly predict the number concentration for cloud droplets and ice crystals (Thompson and Eidhammer, 2014). Therefore, the aerosol-aware Thompson–Eidhammer scheme is an ideal microphysics scheme for evaluating the effect of dust in atmospheric ice nucleation processes. However, this scheme is not coupled with any aerosol model in WRF-Chem. When the aerosol-aware Thompson–Eidhammer microphysics scheme is activated, the model reads in pre-given climatological aerosol data derived from the output of other global climate models, which introduces large errors into the estimation of the effects of dust in microphysical processes. This problem can be solved by embedding a dust scheme into the Thompson–Eidhammer scheme or by coupling the microphysics scheme with WRF-Chem. Compared with WRF, WRF-Chem integrates various emission schemes and aerosol mechanisms for simulating the emission, transport, mixing, and chemical transformation of aerosols simultaneously with the meteorology (Grell et al., 2013). Therefore, WRF-Chem is more capable of producing a realistic aerosol field by comparing the performances of different emission schemes or aerosol mechanisms.
In light of the above, we aim to fully couple the aerosol-aware Thompson–Eidhammer microphysics scheme with the Goddard Chemistry Aerosol Radiation and Transport (GOCART) model (Ginoux et al., 2001) in the WRF-Chem modeling system in this study, enabling WRF-Chem to simultaneously simulate the effect of dust aerosol in ice nucleation processes during simulations. Based upon the implementation, the performance of the coupled GOCART–Thompson microphysics scheme in simulating the ice nucleation process involving dust particles was validated and the role that East Asian dust plays in the ice nucleation process in the atmosphere was further investigated.
The remainder of the paper is presented as follows. Section 2 provides a description of the model, and the implementation work for coupling the aerosol-aware Thompson–Eidhammer microphysics scheme and the GOCART aerosol model in WRF-Chem is elaborated in Sect. 3, followed by the model configurations for numerical simulations in Sect. 4. Section 5 presents the observational data used to validate the performance of the GOCART–Thompson microphysics scheme. Section 6 presents the results and discussion, followed by the conclusions in Sect. 7.
WRF-Chem is an online coupled regional modeling system, which means that it can simultaneously simulate the meteorological field, the chemical field, and the interactions in between (Grell et al., 2013). The chemical model contains several gas- and aerosol-phase chemical schemes. In this study, we focus on the GOCART model, a simple aerosol model that will be used for dust simulation.
GOCART is an aerosol model for simulating major tropospheric natural-source aerosol components, such as sulfate, mineral dust, black carbon, organic carbon, and sea-salt aerosols (Ginoux et al., 2001; Chin et al., 2000). It has been implemented into WRF-Chem as a bulk aerosol scheme. GOCART is a simple aerosol scheme that can predict the mass of aerosol components, but does not account for complex chemical reactions. Therefore, it is numerically efficient in simulating aerosol transport and thus applicable to cases without many chemical processes, especially dust events. Typically, it requires 40 to 50 % more computational time by applying WRF-Chem run with the GOCART aerosol model than the standard WRF to produce the same period of simulation.
Shao's dust emission scheme (Kang et al., 2011; Shao, 2004, 2001; Shao et
al., 2011) is one of the dust emission schemes in the GOCART aerosol model
and has been demonstrated to exhibit superior performance in reproducing the
dust cycle over East Asia compared to other emission schemes (Su and
Fung, 2015). The Shao emission scheme was updated in WRF-Chem since
version 3.8 released in 2016 to produce five size bins for dust emission,
with diameters of
The Thompson microphysics scheme is a bulk two-moment aerosol-aware microphysics scheme that considers the mixing ratios and number concentrations for five water species: cloud water, cloud ice, rain, snow, and a hybrid graupel–hail category (Thompson et al., 2004). The updated Thompson–Eidhammer scheme is an aerosol-aware version of the Thompson scheme (Thompson and Eidhammer, 2014), which incorporates the activation of aerosols serving as cloud condensation nuclei and IN, and therefore it explicitly predicts the number concentrations of CCN and IN, as well as the number concentrations of cloud droplets and ice crystals. Hygroscopic aerosols that serve as cloud condensation nuclei are referred to as water-friendly aerosols, and those non-hygroscopic ice-nucleating aerosols are referred to as ice-friendly aerosols. The cloud droplets nucleate from explicit aerosol number concentrations using a look-up table for the activated fraction as determined by the predicted temperature, vertical velocity, number of available aerosols, and predetermined values of the hygroscopicity parameter and aerosol mean radius.
In the Thompson–Eidhammer scheme, the ice nucleation process is triggered
once the relative humidity with respect to ice (RH
The DeMott parameterization scheme for determining the condensation and
immersion freezing in the Thompson–Eidhammer microphysics scheme was
proposed in 2010 (DeMott et al., 2010, hereafter referred to as the
DeMott2010 scheme) based on combined data from field experiments at a
variety of locations over 14 years. In the Demott2010 parameterization, the
relationship between the number concentration of aerosol-friendly aerosols
and ice-nucleating particles (INP) is as follows:
The above parameterization was further developed in 2015 (DeMott et al.,
2015, hereafter the DeMott2015 scheme) for conditions of a higher mixing ratio
of water vapor or a higher concentration of ice crystals based on the latest
data from field and laboratory experiments. According to the updated
observational data, INP concentration increases exponentially with the number
concentration of ice-friendly aerosols, and existing aerosols with
relatively low concentrations (less than 1000
The number concentration of INP produced by the DeMott2015 scheme is much higher than that produced by the DeMott2010 scheme, and the difference grows larger with decreasing temperature and an increasing number concentration of ice-friendly aerosols (DeMott et al., 2015). Although the DeMott2015 scheme has been implemented in the code of the Thompson–Eidhammer scheme, it cannot be used without modifying the code. Instead of using the DeMott2010 scheme by default, we modified the code to call the DeMott2015 scheme in the Thompson–Eidhammer scheme for the condensation and immersion freezing in our simulations investigate ice nucleation involving dust.
Originally, the calibration factor
Coupling the GOCART aerosol model with the Thompson–Eidhammer microphysics scheme allows the model to explicitly evaluate the indirect effect of natural-source aerosols on the basis of a relatively realistic emission production, for instance the effect of dust on ice nucleation during severe dust episodes or a dust-intensive season.
To investigate the real-time indirect effects of dust aerosol over East Asia, a new treatment was implemented into WRF-Chem to couple the GOCART aerosol model and the Thompson–Eidhammer microphysics scheme, namely the GOCART–Thompson microphysics scheme. To accomplish this, WRF-Chem version 3.8.1 has been modified in the following three steps.
Currently, the GOCART aerosol model generates only the mass concentration for aerosols but no number concentrations. However, the number concentrations of aerosols are required for a microphysics scheme to evaluate the indirect effects of aerosols. Therefore, modification was needed to provide information about the number concentrations of aerosols from the mass concentration produced in the GOCART aerosol model.
The aerosol mass concentration was converted into a number concentration using
the aerosol density and effective radius for each size bin. Assuming that
dust particles are spherical, the mass per dust particle (
This part of the modification was to hook up the GOCART aerosol model and the Thompson–Eidhammer microphysics scheme.
Instead of reading in the pre-given climatological aerosol data, the initialization module of the Thompson–Eidhammer microphysics scheme was modified to apply the bulk number concentration of ice-friendly aerosols produced by the GOCART aerosol model for the calculation of the number concentration of ice-nucleating particles.
After the microphysical processes are finished for a particular time step,
the tendency of the bulk aerosol number concentration (
As no in-cloud scavenging is considered for dust aerosol in WRF-Chem, a new
wet scavenging process was introduced into WRF-Chem to calculate the loss of
aerosol mass due to the microphysical processes within clouds using the
tendency of aerosol number concentration produced by the microphysics
scheme. Assuming that the collection of dust particles is proportional to
the number concentration of dust particles, the fraction of dust particles
for each size bin (
The mass mixing ratio (
Nested domain set for the simulations. Blue dots represent the 10 monitoring stations used for model validation. TD: the Taklimakan Desert; GD: the Gobi Desert.
A numerical experiment was conducted to examine the performance of the
newly implemented GOCART–Thompson microphysics scheme in simulating the ice
nucleation process induced by dust in the atmosphere. Two simulations were
carried out for the numerical test. One control run (CTRL) was simulated
without dust and one test run (DUST) with dust. According to the
observations, the dust events in 2012 over East Asia were concentrated in
mid-March to late April, and the satellite observations from mid-March to
the end of April were available for model validation; therefore, the
simulation period was from 9 March to 30 April 2012, with the first
8 days as “spin-up” time. Only the results from 17 March to 30 April 2012
were used for the analysis. The final reanalysis data provided by the United
States National Center of Environmental Prediction with a horizontal
resolution of 1
Two nested domains were used for the simulations, as shown in Fig. 1. The
outer domain (domain 1) is in a horizontal resolution of 27
In the GOCART–Thompson scheme, deposition nucleation is determined by the Phillips parameterization (Phillips et al., 2008), the freezing of deliquesced aerosols using the hygroscopic aerosol concentration is parameterized following Koop's parameterization scheme (Koop et al., 2000), and the condensation and immersion freezing is parameterized by the DeMott2015 ice nucleation scheme. The new wet scavenging scheme was used for the in-cloud wet scavenging of aerosols due to microphysical processes. The GOCART aerosol model was applied to simulate aerosol processes (Ginoux et al., 2001, 2004) and produce the number concentration of dust particles in DUST. Shao's dust emission (Kang et al., 2011; Shao et al., 2011) with soil data from the United States Geological Survey (USDA, 1993), which have been demonstrated to have good performance in reproducing dust emissions over East Asia, were used to generate dust emissions in the simulations of TEST. The number concentration of dust particles was then fed into the GOCART–Thompson microphysics scheme and treated as ice-friendly aerosols for calculating the condensation and immersion freezing involving dust by the DeMott2015 parameterization scheme. In addition, the pre-given climatological profiles applied in the original Thompson–Eidhammer scheme (Thompson and Eidhammer, 2014) were used to provide the number concentration of water-friendly aerosols for the freezing of deliquesced aerosols calculated by Koops's parameterization scheme to consider the background indirect effect of aerosols on ice nucleation for the simulations of both CTRL and DUST in this study.
Other important physical and chemical parameterizations applied for the simulations are as follows. The Mellor–Yamada–Janjić (MYJ) turbulent kinetic energy scheme was used for the planetary boundary layer parameterization (Janjić, 2002, 1994); the moisture convective processes were parameterized by the Grell–Freitas scheme (Grell and Freitas, 2014); the shortwave (SW) and longwave (LW) radiation budgets were calculated by the Rapid Radiative Transfer Model for General Circulation (RRTMG) SW and LW radiation schemes (Mlawer et al., 1997; Iacono et al., 2008); the gravitational settling and surface deposition were combined for the aerosol dry deposition calculation (Wesely, 1989); a simple washout method was used for the below-cloud wet deposition of aerosols (Duce et al., 1991; Hsu et al., 2006); and the aerosol optical properties were calculated based on the volume-averaging method (Horvath, 1998).
Hourly observations of the surface concentration of particulate matter with
a diameter smaller than 10
The AERONET program is a ground-based aerosol remote sensing network for
measuring aerosol optical properties at sites distributed around the globe.
This program provides a long-term database of aerosol optical properties
such as aerosol extinction coefficient, single-scattering albedo, and
aerosol optical depth (AOD) measured at various wavelengths. The
observational data from two sites were available for comparison with the
simulation results during the simulation period in this study. One was
Dalanzadgad, located to the north of the Gobi Desert in Mongolia, and the
other was the Semi-Arid Climate and Environment Observatory of Lanzhou
University (SACOL), located in Lanzhou, Gansu Province, China. The exact
locations of the two AERONET sites are depicted by the red triangles in
Fig. 1. All of the measured data passed the quality control standard
level 2 with an uncertainty of
The MISR instrument aboard the Terra platform of the United State National
Aeronautics and Space Administration (NASA) has been monitoring aerosol
properties globally since 2000. It measures the aerosol properties in four
narrow spectral bands centered at 443, 555, 670, and 865
The MODIS instruments aboard the Terra and Aqua platforms of NASA monitor Earth's surface and provide global high-resolution cloud and aerosol optical properties at a near-daily interval (Kaufman et al., 1997).
To retrieve aerosol information over bright surfaces, the Deep Blue
algorithm was developed to employ retrievals from the blue channels of the
MODIS instruments, at which wavelength the surface reflectance is very low
such that the presence of aerosol can be detected by increasing total
reflectance and enhanced spectral contrast (Hsu et al., 2006). By
applying this algorithm, the AOD values at wavelengths of 214, 470,
550, and 670
The Cloud-Aerosol Lidar and Infrared Pathfinder Satellite, which is aboard
the Aqua platform of NASA, combines an active light detection and ranging
(lidar) instrument with passive infrared and visible imagers to probe the
vertical structure and properties of thin clouds and aerosols around the
globe (Vaughan et al., 2004). It aims to fill existing gaps in the
ability to measure the global distribution of aerosols and cloud properties
and provides three-dimensional perspectives of how clouds and aerosols form,
evolve, and affect weather and climate. It measures high-resolution vertical
profiles of aerosol and cloud extinction coefficients globally at wavelengths
of 532 and 1064
Time series of spatially averaged daily dust mass load
The time series of daily average dust load over the entire East Asia region
(domain 1) during the simulation period is shown in Fig. 2a. In total four
dust events occurred during the simulation period, lasting from 18 to
25 March, 30 March to 7 April, 9 to 19 April, and 22 to 29 April 2012. The case
from 22 to 29 April was the most significant one, with a daily dust load
double that of the other cases. The fraction of daily dust load for each size bin
is also shown in Fig. 2a. The dust particles in the fourth and fifth bins
with effective diameters ranging from 6 to 20
The number concentrations of dust particles over East Asia were vertically
integrated to obtain the number density of dust particles. As shown in
Fig. 2b, the time series of the daily average number density of dust
particles over East Asia during the simulation period shows a similar
distribution as that for dust load; the noteworthy distinction between the
two time series lies in the fraction of each size bin. The two size bins
with the smallest diameters (no larger than 3.6
Time series of hourly observed and simulated surface PM
Performance statistics for the model in
simulating surface PM
MB: mean bias; ME: mean error; RMSE: root mean squared error;
To evaluate the performance of WRF-Chem in reproducing dust emissions over
East Asia, the simulated surface PM
The performance statistics were computed from the daily average simulated
PM
Time series of daily mean observed and simulated aerosol optical
depths at Dalanzadgad
To examine the performance of the model in reproducing the column sum of dust in the atmosphere, the simulated AOD values were compared with observations measured at two AERONET sites during the simulation period, as shown in Fig. 4.
The site at Dalanzadgad (Fig. 4a) is located in Mongolia to the north of the Gobi Desert. Overall, the evolution and magnitude of the AOD time series at Dalanzadgad were reasonably reproduced by the model during the simulation period, despite the fact that the simulated AOD was overestimated at the end of March and in mid-April compared to the observed values.
SACOL (Fig. 4b) is a site located in Lanzhou, Gansu Province, which is a typical downwind area for dust in China. The model showed a good performance in reproducing the time series of AOD at SACOL during the entire simulation period, with the evolution and magnitude of AOD well captured.
Spatial distributions of monthly mean AOD from MODIS observations
The spatial distribution of monthly mean simulated AOD is also compared with observed values from MODIS and MISR products in Fig. 5. Note that the high AOD values observed in northern, eastern, and southern China and part of Southeast Asia are attributed to the abundant anthropogenic emissions, while those high values in the circle area are mostly due to dust events. The region with high AOD values in the western part of the circled area is TD, and the region with relatively lower AOD in the eastern part of the circled area is GD. The AOD observed by MODIS showed high values at the dust source region in both March and April of 2012, as shown in Fig. 5a and b. The mean observed AOD over GD was lower than that over TD in both March and April, and the mean observed AOD was higher in April than in March over both dust source areas. The spatial patterns of AOD observed by MISR are similar to MODIS, with comparable mean values over GD. However, the mean AOD values over TD observed by MISR are 36 and 40 % lower than those by MODIS in March and April, respectively (Fig. 5c and d).
The spatial patterns for the mean simulated AOD were similar to the observed values in both months but closer to those from MODIS, as shown in Fig. 5e and f. The model shows a good capability in capturing the spatial characteristics of the AOD over the dust source areas. For example, the mean observed AOD was higher in the southern part of TD than that in the northern part in March and showed an increase from March to April over GD, both of which were captured by the model. The values of the mean simulated AOD over the Gobi Desert (0.33 for March and 0.39 for April) are comparable to the observational values from both MODIS (0.30 for March and 0.32 for April) and MISR (0.31 for March and 0.34 for April), but the mean simulated AOD values over TD (0.54 for March and 0.64 for April) are between the values of the MISR observations (0.72 for March and 0.88 for April) and the MODIS observations (0.46 for March and 0.53 for April).
2-D histogram of simulated cloud ice mixing ratio
Spatial distributions for the temporal mean simulated cloud ice
water path
In summary, it was demonstrated that the dust emissions simulated by
WRF-Chem are reliable for further analysis by the comparison between the
simulation results and the observations for surface PM
Dust particles are effective IN and play an important role in ice nucleation in the atmosphere under appropriate conditions. With the large number of IN created by dust particles emitted into the atmosphere, an increase in the number of ice crystals is expected in the results from DUST compared with those from CTRL, after taking into account the effects of dust particles in the GOCART–Thompson microphysics scheme. Figure 6 shows the overall comparison between the number of grid points of simulated cloud ice mixing ratio and ice crystal number concentration in corresponding value bins (at all model grids at hourly intervals) from CTRL and DUST during the entire simulation period.
As expected, the model produces a higher cloud ice mixing ratio (Fig. 6a)
and ice crystal number concentration (Fig. 6b) in DUST. The simulated
cloud ice mixing ratio produced in DUST is substantially higher than that
produced in CTRL throughout all value bins, especially in those bins with
values lower than 0.05
The spatial distributions of the simulated IWP and ice crystal number
density from CTRL and DUST in Fig. 7 further demonstrate the enhancement
in cloud ice due to dust over East Asia. The IWP produced by CTRL was
relatively high over western and eastern China, as well as at the southern boundary
of the simulation domain, with values as high as 20
The mean IWP and ice crystal number density were increased by 15 and 8 % over vast areas of East Asia upon considering the effect of dust in the ice nucleation process in the simulation, and such an effect can reach as far as the open ocean of the Western Pacific (Fig. 7b and e), as the outbreak of a cold high system over northeast Asia can bring quantitative dust aerosol down to the Western Pacific or even further during the dust season.
Spatial distribution for simulated dust load and satellite
scanning track
As Fig. 8 but for the cases on 9 April (left column) and 23 April (right column) of 2012.
The vertical profile of the simulated IWC was also compared with the observation from CALIPSO during dust events. As mentioned in Sect. 5.1, a total of four dust events occurred during the simulation period, lasting from 18 to 25 March, 30 March to 7 April, 9 to 19 April, and 22 to 28 April 2012. As shown in Figs. 8 and 9, the performance of the model in simulating the vertical profile of IWC was evaluated by comparing the observations measured at 06:00 UTC on 21 March, 18:00 UTC on 1 April, 18:00 UTC on 9 April, and 05:00 UTC on 23 April 2012 with the simulated profiles at the same hour.
CALIPSO measures the global distribution of aerosol and cloud properties by
lidar, which uses a laser to generate visible light with a wavelength of 1
The simulated dust load over East Asia at 06:00 UTC on 21 March 2012 is shown
in Fig. 8a, in which the dust covered vast areas from western to eastern China
between 35 and 45
On 1 April 2012, central to eastern China was covered by a thick dust plume,
and the orbit of the satellite passed between 25 and
43
Vertical profiles for the mean observed IWC from CALIPSO and the simulated IWC from CTRL and DUST for dust events on 21 March and 1, 9, and 23 April 2012.
At 18:00 UTC on 9 April 2012, the satellite was scanning the dust source over
GD, which was covered by a thick dust plume between 35 and
45
Similar to the previous cases, the satellite was scanning along the east coast
of China at 05:00 UTC on 23 April 2012, when a dust plume was arriving
from the dust sources and affecting areas between 35 and
45
By comparing the satellite-observed and simulated vertical profiles of IWC during the various dust events, it was demonstrated that the model reproduces the enhancement of IWC clouds in the middle to upper troposphere by taking into account the effect of dust in the ice nucleation process, which substantially improves the simulation of cloud ice.
The mean profiles of the observed IWC, as well as the simulated IWC from
CTRL and DUST for the four dust events discussed in Sect. 6.2.2, are shown
in Fig. 10. Note that the “mean profile” of IWC is the average over the
available data points for the IWC along the orbit of the satellite between
30 to 45
Compared with the results from CTRL, the vertical profile of the simulated IWC was substantially improved in DUST for each dust event, with the enhancement of the ice nucleation process well captured by the GOCART–Thompson microphysics scheme. However, there were still discrepancies between observations and the simulation results from DUST, and the magnitudes of the vertical IWC produced by the model were always lower than the observed values.
For the cases on 21 March and 1 April, the peaks of IWC were observed at 9.5 and 8
As discussed in Sect. 6.2.3, the simulation of cloud ice is greatly improved by considering the enhancement of the ice nucleation process induced by dust, which is well captured by the GOCART–Thompson microphysics scheme. However, the IWC is still underestimated by the model during dust events. To determine the reason for this limitation, numerical experiments were performed to investigate the sensitivity of simulated IWC to the parameters of the ice nucleation parameterization in the GOCART–Thompson microphysics scheme.
Vertical profiles for the mean observed IWC from CALIPSO and the
simulated IWC with various
The calibration factor
The mean profiles of IWC from simulation results were compared with the
CALIPSO observations for the dust events discussed in Sect. 6.2.2 and
6.2.3, as shown in Fig. 11. For the cases on 21 March and 1 April,
changing
For the case on 9 April, the simulated IWC increased between 6 and
9
For the case on 23 April, two peaks were observed in the profiles of
simulated IWC located at 7 and 10
In summary, increasing the calibration factor
Vertical profiles for the mean observational IWC from CALIPSO and the simulated IWC with threshold RH values of 105 and 100 % for the dust events on 21 March and 1, 9, and 23 April 2012.
As ice nucleation occurs only in a supersaturated atmosphere with respect
to water vapor, the ice nucleation process would be terminated in the
GOCART–Thompson microphysics scheme when the environmental RH
In this study, the threshold relative humidity to trigger the ice nucleation process in the simulation was originally set to 105 %, which was selected for the central lamina condition in the laboratory experiments to derive the DeMott2015 ice nucleation scheme (DeMott et al., 2015). However, as reported in other studies, the number of ice-nucleating particles starts to rise when the relative humidity exceeds 100 % (DeMott et al., 2011). Therefore, a sensitivity experiment was carried out to investigate the response of simulated IWC to a lower threshold relative humidity.
The mean profiles of IWC from the simulation results were compared with the CALIPSO observations for the aforementioned dust events, as shown in Fig. 12. With the threshold relative humidity lowered to 100 %, the simulated IWC showed an increase throughout the vertical profile with the most significant increase at the peaks, suggesting more consistency with the observations for all of the dust events, except the one on 1 April. In the case on 1 April, the simulated IWC increased at lower altitudes than the observed peak, but slightly decreased right at the peak with lowering the threshold relative humidity to 100 %. Overall, the simulation of IWC during dust events was significantly improved by lowering the threshold relative humidity from 105 to 100 %.
A new treatment, the GOCART–Thompson scheme, was implemented into WRF-Chem to couple the GOCART aerosol model to the aerosol-aware Thompson–Eidhammer microphysics scheme. By applying this newly implemented microphysics scheme, the effect of dust on the ice nucleation process by serving as IN in the atmosphere can be quantified and evaluated. Numerical experiments, including a control run without dust and a test run with dust, were then carried out to evaluate the performance of the newly implemented GOCART–Thompson microphysics scheme in simulating the effect of dust on the content of cloud ice over East Asia during a typical dust-intensive period by comparing the simulation results with various observations.
Based on the GOCART aerosol model the model reproduced dust emissions
reasonably well by capturing the evolution and magnitude of the surface
PM
The effect of dust on the ice nucleation process was then quantified and evaluated in the GOCART–Thompson microphysics scheme. Upon considering the effect of dust in the simulation, the simulated ice water mixing ratio and ice crystal number concentration over East Asia were 15 and 7 % higher than those simulated without dust, with the most significant enhancements located over dust source regions and downwind areas.
Comparison between the vertical profiles of the satellite-observed and simulated IWC during various dust events indicated that the enhancement of cloud ice induced by abundant dust particles serving as IN is well captured by the GOCART–Thompson microphysics scheme, with the results from the simulation with dust much more consistent with the satellite observations, although the IWC is generally underestimated by the model.
Sensitivity experiments revealed that the simulated IWC was not very sensitive to the calibration factor defined in the DeMott2015 ice nucleation scheme, but the model delivered a slightly better performance in reproducing the IWC when the calibration factor was set to 5. However, the simulated IWC was sensitive to the threshold relative humidity to trigger the ice nucleation process in the model and was improved by lowering the threshold relative humidity from 105 to 100 %.
The WRF-Chem outputs and surface PM
The authors declare that they have no conflict of interest.
We would like to acknowledge the provision of the MODIS and the MISR observations by the Ministry of Environmental Protection Data Center, U.S. National Center for Atmospheric Research (NCAR) and the CALIPSO data by the U.S. National Aeronautics and Space Administration (NASA) Data Center. We thank the principal investigators and their staff for establishing and maintaining the two AERONET sites used in this study. We appreciate the assistance of the Hong Kong Observatory (HKO), which provided the meteorological data. Lin Su would like to thank Georg Grell, Stuart McKeen, and Ravan Ahmandov from the Earth System Research Laboratory, U.S. National Oceanic and Atmospheric Administration for insightful discussions. Other data used this paper are properly cited and referred to in the reference list. All data shown in the results are available upon request. This work was supported by NSFC/RGC grant no. HKUST631/05, NSFC-FD grant U1033001, and RGC grant 16303416. Edited by: Corinna Hoose Reviewed by: Gregory Thompson and two anonymous referees