1. Introduction
With a mean elevation of about 4000 m, the Tibetan Plateau (TP), also referred to as the “roof of the world,” is situated in the interior of the Eurasian continent. In addition, the TP is also called the “Water Tower of Asia” and the “Third Pole,” playing an essential function in global climate and hydroecology (Xu et al. 2014; Yao et al. 2019; Y. Liu et al. 2022). Several major dust sources exist around the TP, mainly including the Taklamakan Desert and the Gobi Desert in East Asia, the Thar Desert in South Asia, the Karakum Desert in Central Asia, the Rub Al Khal Desert and the Al Nafud Desert in the Middle East, and the Sahara Desert in North Africa (Mao et al. 2019; Hu et al. 2020; Zhu et al. 2021). Because of these deserts, dust is the primary category of aerosol over the TP and peaks during spring (Huang et al. 2007, 2008; C. Xu et al. 2020). Analysis of glaciers and snow cover revealed the presence of dust particles over the TP (Wu et al. 2010; Kang et al. 2010; Li et al. 2019; Dong et al. 2020), and subsequent chemical testing revealed that the dust layers in the ice and snow over TP are created by the deposition of dust particles that had traveled great distances (Osada et al. 2004; Xu et al. 2012; Dong et al. 2009, 2014).
Using satellite-based lidar, previous studies have shown that the dust originating from the Taklamakan Desert can be lifted to the northern slope of the TP (Huang et al. 2007; Y. Liu et al. 2019; X. Xu et al. 2020). Combining multiyear satellite measurements and trajectory models, Jia et al. (2015) discovered that the Taklamakan Desert is the primary source of the dust aerosols over the TP, in addition, the Gurbantunggut Desert and Thar Desert also make some contributions. Recently, based on satellite data and reanalysis products, the transport of dust from Central Asia and South Asia to the TP has been identified and quantified (Li et al. 2020; T. Wang et al. 2021; Han et al. 2022). Furthermore, results based on in situ observations also indicate that dust originating from the surrounding areas of the TP can be transported to the TP (Che et al. 2015; Zhu et al. 2019; Zhang et al. 2021). For more distant dust sources, based on observations and reanalysis data, Lau and Kim (2006) and Lau et al. (2018) reported that dust aerosols that are transported from the Middle East mainly accumulate over the southern slope of the TP. Moreover, using satellite observations and numerical models, Q. Liu et al. (2022) found that the dust from Sahara Desert can reach the TP over long distances. In addition, some numerical simulation research has shown that dust from several major global dust sources can all contribute to the dust over the TP (Chen et al. 2013; Feng et al. 2020; Zhu et al. 2021).
Recent studies have begun to measure the contribution of different dust sources to the dust over the TP. Mao et al. (2019) estimated the contributions of dust from East Asia, the Middle East, and Central Asia as well as North Africa to the dust over the TP based on a simulation during 2010–14, they pointed the concentration of dust over the TP is the highest in spring, with East Asian and North African dust contributing about 60% and 22%, respectively, and dust from Middle East and Central Asia contributing about 18%. Hu et al. (2020) also used a numerical model to study the transport and proportion of several dust sources to dust over the TP at different altitudes from 2010 to 2015 and further analyzed their radiative effects. These results quantitatively indicate that dust over the TP is influenced by multiple sources. However, these studies only analyzed the proportion of dust from different sources in the average dust concentration over the TP, but did not involve the impact of different sources on dust variability over the TP, and longer-term numerical simulations are still needed to obtain more accurate results. In addition, the mechanisms associated with the impact of different sources on dust variability over the TP also need to be analyzed.
Accordingly, by performing a 9-yr simulation and combining data analysis, this study quantitatively shows the relative contributions made by five major dust sources to the distributions and variations of spring dust over the TP and reveals the corresponding mechanisms. This essay has the following organizational structure. The data, model, and methods employed in this study are presented in section 2 after the introduction in section 1. The results acquired from satellite observations, reanalysis data, and numerical simulations are shown in section 3. Finally, section 4 provides a succinct summary of the findings.
2. Materials and methods
a. Observations and reanalysis data
The Multiangle Imaging SpectroRadiometer (MISR) aboard the Terra satellite provides radiance, aerosol, land surface, albedo, and cloud observations in four spectral bands and eight angles of visible and near-infrared light since 2000. The MISR Level 3 global monthly aerosol optical depth (AOD) product (Martonchik et al. 2002), with a resolution of 0.5° × 0.5°, is used in this study to analyze the distributions of AOD and to evaluate the AOD from reanalysis data for the 2000–20 period. In addition, monthly averaged sea surface temperature (SST) with a resolution of 1° × 1° from Hadley Centre Sea Ice and Sea Surface Temperature dataset (HadISST; Rayner et al. 2003) during 2000–20 is used to calculate the northern North Atlantic SST (41°–62°N, 10°–52°W).
The Modern-Era Retrospective Analysis for Research and Applications version 2 (MERRA-2) is a reanalysis data produced by NASA’s Global Modeling and Assimilation Office (GMAO), covering the period since 1980 (Gelaro et al. 2017), which includes assimilations of AOD from multiple ground-based observation and space-based remote sensing systems (Randles et al. 2017). MERRA-2 uses the GEOS-5 model coupled with the GOCART model to correctly simulate procedures associated to aerosols, including radiation and microphysics, emissions, transport, and dry and wet deposition of black carbon, organic carbon, dust, sulfate, and sea salt aerosols. These data have been widely utilized in studies relating to aerosols and are applicable through comparison with satellite and site observations (L. Liu et al. 2019; Sun et al. 2019a,b; Liu et al. 2023). This study applied the monthly AOD, dust column mass density, and column mass flux from 2000 to 2020 provided by MERRA-2 aerosol diagnostic product (tavgM_2d_aer_Nx) to analyze the characteristic and variation of dust as well as to evaluate simulations. The horizontal resolution of this product is 0.625° × 0.5°.
Additionally, the monthly geopotential height, vertical velocity, and zonal and meridional winds for the period from 2000 to 2020 obtained from ERA5 reanalysis data (Hersbach et al. 2020) are employed to analyze the characteristics of atmospheric circulations and to calculate the Rossby wave activity flux (Takaya and Nakamura 2001), with a spatial resolution of 0.5° × 0.5° in the horizontal scale.
b. Model and experiment description
The Nonhydrostatic Icosahedral Atmospheric Model (NICAM) coupled online with the Spectral Radiation-Transport Model for Aerosol Species (SPRINTARS) (Satoh et al. 2014; Suzuki et al. 2008; Tomita and Satoh 2004) is used to quantify the contributions of different dust sources to the dust over the TP and to analyze relevant mechanisms. The NICAM is a global nonhydrostatic model. In this study, we perform the global simulation with a horizontal resolution of 56 km and a vertical resolution of 40 layers from the surface to approximately 40-km altitude. As a global 3D aerosol radiation transport model, the SPRINTARS has been reconciled with NICAM, five kinds of aerosols are considered in the coupled NICAM-SPRINTARS, namely sulfate, black carbon, organic carbon, sea salt, and dust (Dai et al. 2015; Takemura et al. 2000).
Distribution of defined regions in NICAM-SPRINTARS.
Citation: Journal of Climate 37, 10; 10.1175/JCLI-D-23-0486.1
To assess the accuracy of NICAM, the control experiment was conducted using the standard dust emission coefficients. Additionally, the individual dust emissions from North Africa (NA), the Middle East (ME), South Asia (SA), Central Asia (CA), and East Asia (EA) are obtained by only keeping the dust emission coefficients of region 2, 4, 5, 6, or 7, respectively, and setting the dust emission coefficients of other regions to 0 (Fig. 2). These sensitivity experiments are used to quantify the contributions of different sources to the dust over the TP. The 6-h final analysis (FNL) dataset is used for the initial conditions and meteorological nudging, including the wind, pressure, temperature, and specific humidity nudging with a time scale of 6 h. According to the time series of the dust optical depth (DOD) anomaly over the TP derived from MERRA-2 reanalysis data (Fig. 3e), three years with the highest values are selected as the strong DOD years (SDY; 2008, 2012, and 2018), three years with the lowest values are selected as the low DOD years (LDY; 2014, 2015, and 2016), and three years closest to the average value are selected as the average DOD years (2006, 2001, and 2011). The simulations are output day by day and focus on the spring (1 March–31 May) of the above selected 9 years, and 15 days before 1 March of each year are discarded as the spinup period to lessen the effect of initial conditions.
Regions of (a1) NA, (b1) ME, (c1) SA, (d1) CA and (e1) EA. (a2)–(e2) Spatial distributions of the average dust emission flux (μg m−2 s−1) of the corresponding regions during the simulation period. The major body of the TP is marked by the region surrounded by the purple bold curve.
Citation: Journal of Climate 37, 10; 10.1175/JCLI-D-23-0486.1
Spatial distributions of the average AOD at 550 nm (a) derived from MISR and (b) from MERRA-2 during the springs of 2000–20. (c) Scatterplot of spring AODs over the TP between MERRA-2 and MISR during 2000–20. (d) Spatial distributions of the average DOD derived from MERRA-2 during the springs of 2000–20. (e) Time series of DOD anomaly over the TP derived from MERRA-2 during the springs of 2000–20.
Citation: Journal of Climate 37, 10; 10.1175/JCLI-D-23-0486.1
3. Results
a. Distribution and variation of dust over the TP
The MERRA-2 reanalysis product is a powerful tool for evaluating dust distributions and variations since it can provide characteristics of five types of aerosols, including dust (Feng et al. 2020; Lau et al. 2018; M. Wang et al. 2021). To determine the applicability of aerosol products from MERRA-2 over the TP, the AOD over the TP from MERRA-2 is compared with that from MISR observations. Figures 3a and 3b, respectively, depict the spatial distributions of the average AODs from MERRA-2 and MISR for the springs of 2000–20. Both the distributions of AOD from MERRA-2 and MISR are characterized by high AODs over the edge of the TP and Tsaidam Basin as well as characterized by low AODs throughout the main body of the TP. In addition, the correlation between the time series of AODs over the TP from MERRA-2 and MISR during the springs of 2000–20 is positive and significant (Fig. 3c), demonstrating a respectably high level of agreement between MERRA-2 and MISR AODs. Therefore, based on MERRA-2 reanalysis data, the spatial distribution of dust optical depth (DOD) over the TP during the springs of 2000–20 is shown in Fig. 3d. Dust is primarily dispersed across the edge of the TP and Tsaidam Basin, especially over the northern slope of the TP, which is consistent with the distribution of aerosol. The dust contributes to 51.91% of the AOD over the TP, which is the main aerosol type, and the time series of the DOD anomaly over the TP is shown in Fig. 3e.
According to the distributions of DOD and dust column mass flux in Fig. 4a, the TP is surrounded by multiple regions with high DOD, including deserts in NA, ME, SA, CA, and EA. In addition, the dust from NA, ME, SA, and CA is transported toward the TP with westerlies, and the dust from EA enters the TP via the northern slope. Furthermore, the correlations between DODs over different sources and DOD over the TP are positive and significant (Figs. 4b–f). Based on the time series of the DOD anomaly over the TP (Fig. 3e), the years with the top 25% highest values are selected as the high DOD years (2008, 2012, 2018, 2007, and 2010), and the anomaly field of the dust column mass flux is calculated for the high DOD over the TP relative to the average field from 2000 to 2020. As shown in Fig. 4g, when the DOD over the TP is greater, an exceptionally high dust column mass flux is found over North Africa to East Asia. The transport of dust from NA, ME, SA, and CA to the TP with westerlies is boosted and that from EA southward into the TP is also enhanced. The distribution and the anomaly field of dust column mass flux, as well as the correlations between DODs over different sources and DOD over the TP, indicate that all of these dust sources may be involved in the distribution and the variation of dust over the TP.
(a) Spatial distribution of the average DOD (shading) and dust column mass flux (vectors; g m−1 s−1) during the springs of 2000–20. The blue, yellow, green, pink, and red rectangles represent the areas for calculating DOD over NA, ME, SA, CA, and EA, respectively. The major body of the TP is marked by the region surrounded by the purple bold curve. (b)–(f) Scatterplots of DOD over NA, ME, SA, CA, EA, and DOD over the TP during the springs of 2000–20, respectively. (g) Spatial distribution of the anomaly field of dust column mass flux (g m−1 s−1) with increased DOD over the TP relative to the average field in spring from 2000 to 2020.
Citation: Journal of Climate 37, 10; 10.1175/JCLI-D-23-0486.1
b. Transport and contributions of different sources to dust over the TP
To further distinguish the contributions of different sources to dust over the TP, six groups of NICAM-SPRINTARS experiments are performed (the details of these experiments are described in section 2b). Figures 5a and 5b show the average DOD derived from MERRA-2 and the control experiment of NICAM-SPRINTARS, respectively, which indicate the spatial distribution of DOD simulated by NICAM-SPRINTARS agrees well with that derived from MERRA-2; they all show a dust belt from NA to EA, and high values of DOD occur over the main sources. Figure 5c presents the comparison of DOD between the simulations under the control experiment and MERRA-2 over different regions, which demonstrates that besides giving appropriate patterns (Figs. 5a,b), NICAM-SPRINTARS can accurately measure the values of DOD. The mean deviation of simulations from MERRA-2 is 5.01% (the maximum is 7.60% over SA, and the minimum is 1.15% over CA). Therefore, NICAM-SPRINTARS is overall deemed reliable for investigating the values and distributions of dust over the study area.
Distributions of DOD from (a) MERRA-2 and from (b) NICAM-SPRINTARS in the spring of nine selected years. (c) Mean DOD from MERRA-2 product and NICAM-SPRINTARS simulations over different regions during the spring of nine selected years.
Citation: Journal of Climate 37, 10; 10.1175/JCLI-D-23-0486.1
To more accurately illustrate the impact of different sources on dust over the TP, the spatial distributions of annual mean spring dust concentrations released from NA (Figs. 6a1–a3), ME (Figs. 6b1–b3), SA (Figs. 6c1–c3), CA (Figs. 6d1–d3), and EA (Figs. 6e1–e3), as well as wind vectors at different layers, are illustrated. At 600 hPa, the dust originating from NA is transported eastward with westerlies and accumulates over the western and southern sides of the TP (Fig. 6a1). At 400 hPa, the dust originating from NA can rise to the TP and is mainly distributed over the western and southern slopes of the TP (Fig. 6a2). However, at 250 hPa, the western and northern slopes of the TP are where most of the dust from NA is spread (Fig. 6a3). Similar to the transport characteristic of the dust from NA, the transport characteristic of the dust from ME to the TP is shown as the dust transport to the TP with westerlies and gathered over the western and southern sides of the TP at the lower and middle layers, and concentrated over the western and northern slopes of the TP at the upper layers (Figs. 6b1–b3). The South Asian dust is converged over the southern slope of the TP at all vertical levels (Figs. 6c1–c3). Conversely, the northern slope is where the majority of the dust that CA emits enters the TP (Figs. 6d1–d3). Moreover, according to Figs. 6e1–e3, the dust from EA predominates in the dust over the northern TP and can approach the southern TP.
Distributions of annual mean spring wind vectors (m s−1) and dust concentrations (μg m−3) emitted from (a1)–(a3) NA, (b1)–(b3) ME, (c1)–(c3) SA, (d1)–(d3) CA, and (e1)–(e3) EA at (left) 600, (center) 400, and (right) 250 hPa simulated by NICAM-SPRINTARS during the spring of nine selected years.
Citation: Journal of Climate 37, 10; 10.1175/JCLI-D-23-0486.1
Additionally, the vertical distributions of dust from different sources to the TP are analyzed. Figure 7 shows the altitude–longitude cross sections of dust concentration along 30° and 35°N. Due to the long distance of NA and ME from the TP, the dust originating from these sources can be advected into the TP after reaching high altitude, and can also upslope to the TP along the terrain with strong updrafts when reaching the western and southern slopes of the TP (Figs. 7a1,b2). Comparing Figs. 7c1 and 7c2, it is found that the dust concentration over the TP along 30°N is significantly higher than that along 35°N, indicating dust emitted from SA reaches the TP primarily through the southern slope of the TP. Similarly, the comparison of Figs. 7d1 and 7d2 indicates that the dust released by CA enters the TP mainly through the western slope of the TP. Furthermore, the dust from EA is distributed over the northern and eastern slopes of the TP (Figs. 7e1,e2).
The mean dust concentration (shading; μg m−3) emitted from NA and vertical circulation (vectors; m s−1) along (a1) 35° and (a2) 30°N derived from NICAM-SPRINTARS in the spring of nine selected years. (b1),(b2), (c1),(c2), (d1),(d2), (e1),(e2) As in (a1) and (a2), but for dust concentrations released from ME, SA, CA, and EA, respectively. The vertical velocity has been increased by 700 times to make it equivalent to the horizontal wind.
Citation: Journal of Climate 37, 10; 10.1175/JCLI-D-23-0486.1
Figure 8 depicts the latitude–height sections along 75°, 82.5°, and 95°E. According to Figs. 8a1–b3, the concentration of dust transported to the TP is greatest along 75°E, which implies that the dust originating from NA and ME is transported to the TP primarily via the western slope of the TP. The comparison of Figs. 8c1–c3 shows that the dust concentration over the TP is highest along 82.5°E, indicating the dust emitted from SA reaches the TP mostly through the southern slope of the TP. Concentration of dust transported from CA to the TP is highest along 75°E, and the dust concentration is the highest over the northern slope of the TP (Fig. 8d1), indicating the dust originating from CA enters the TP mainly through the northern and western slope of the TP. Moreover, the East Asian dust concentration along 82.5°E is more severe than that along 75° and 95°E (Figs. 8e1–e3), implying the East Asian dust enters the TP with strong updrafts through the northern slope of the TP. These above results suggest the entry of dust into the TP is related to the midlatitude westerlies and vertical ascending motion over the edge of the TP.
The mean concentration (shading; μg m−3) of the dust emitted from NA and vertical circulation (vectors; m s−1) along (a1) 75°, (a2) 82.5°, and (a3) 95°E simulated by NICAM-SPRINTARS during the spring of nine selected years. (b1)–(b3),(c1)–(c3),(d1)–(d3),(e1)–(e3) As in (a1)–(a3), but for dust concentrations emitted from ME, SA, CA, and EA, respectively. The vertical velocity has been increased by 700 times.
Citation: Journal of Climate 37, 10; 10.1175/JCLI-D-23-0486.1
The column average dust concentration over the TP indicates that the dust is primarily distributed on the northern slope of the TP (Fig. 9a), which is mainly contributed by the East Asian deserts (Figs. 9f and 10e). North African and Middle Eastern dust transported to the TP is primarily distributed on the western and southern slopes of the TP (Figs. 9b,c), with Middle Eastern dust contributing the most to the western and southern parts of the TP (Fig. 10b), and North African dust also makes up a significant proportion of the dust over the western part of the TP (Fig. 10a). The dust originating from SA and CA transported to the TP is distributed over the southern and western slopes of the TP, respectively (Figs. 9d,e). Additionally, dust from NA, ME, and EA practically has impacts throughout the whole TP (Figs. 10a,b,e), but dust from SA and CA only has influence on parts of the TP (Figs. 10c,d).
(a) Distribution of annual mean spring column average dust concentration (μg m−3) over the TP in the control experiment during the spring of nine selected years. Distribution of annual mean spring column average dust concentration (μg m−3) over the TP contributed by dust sources over (b) NA, (c) ME, (d) SA, (e) CA, and (f) EA simulated by NICAM-SPRINTARS during the spring of nine selected years.
Citation: Journal of Climate 37, 10; 10.1175/JCLI-D-23-0486.1
Distribution of annual mean spring column average dust concentration fractions (%) over the TP contributed by dust sources over (a) NA, (b) ME, (c) SA, (d) CA, and (e) EA simulated by NICAM-SPRINTARS during the spring of nine selected years.
Citation: Journal of Climate 37, 10; 10.1175/JCLI-D-23-0486.1
Figure 11a displays the column concentration percentages (%) over the TP of dust originating from different sources, which indicates that East Asian dust is the paramount contributor to the spring dust concentration over the TP, accounting for 42%–68% of the total dust concentration. Moreover, dust from ME and NA also makes a certain contribution, making up 13%–27% and 9%–25% of the total dust concentration, respectively. Only a small percentage of the total dust concentration (3%–10% and 1%–8%, respectively) is made up of dust from SA and CA.
(a) The maximum, average, and minimum column average dust concentration percentage (%) of different dust sources over the TP simulated by NICAM-SPRINTARS during the spring of selected 9 years. (b) As in (a) but for average dust concentration percentage (%) at the vertical layers. (c) Purple bars indicate column average dust concentration over the TP (μg m−3) simulated by NICAM-SPRINTARS during the SDY and the LDY, and their difference (ΔD). Orange bars indicate the concentration percentage (%) of dust from different dust sources to ΔD.
Citation: Journal of Climate 37, 10; 10.1175/JCLI-D-23-0486.1
The average concentration percentages of dust from different sources in different vertical layers over the TP are depicted in Fig. 11b. The percentage of dust from EA is the highest (the red line) below 400 hPa, accompanied by that from ME (the yellow line) and NA (the blue line), and the percentage of dust from CA is the smallest (the pink line). However, above 350 hPa, the percentage of the Middle Eastern dust is the largest, followed by those of dust from NA and EA. The percentages of average dust concentration over the TP from different sources in various layers are substantially diverse: the percentage of East Asian dust is the largest (66.08%) at 600 hPa, while the percentage plunges to 21% at 200 hPa, which decreases with increasing altitude. Conversely, the percentages of dust originating from ME and NA are 17% and 11% at 600 hPa, respectively, and those grow with increasing altitude, reaching 37% and 34% at 200 hPa, respectively. These variations are caused by atmospheric circulations and the different distances between different sources and the TP. As the East Asian deserts are closest to the TP (Fig. 2e2), dust originating from EA is more easily transported to the TP, which leads to the majority of the dust over the TP in the lower and middle layers is contributed by the East Asian desert. However, when dust from EA is transported to the TP, it is also transported eastward under the influence of westerlies, which makes East Asian dust gradually transport to the downstream regions of the TP (Figs. 6e1–e3 and 7e1,e2), causing a decrease in the proportion of East Asian dust at upper layers. Conversely, the deserts over the ME and NA are far from the TP and the dust from these sources can be elevated before approaching the TP, and then advected into the TP by westerlies in the middle and upper layers (Figs. 6a1–b3 and 7a1–b2), which causes greater proportions of Middle Eastern and North African dust in the upper layers.
Moreover, the simulated column average dust concentration over the TP during the SDY (2008, 2012 and 2018) and the LDY (2014, 2015, and 2016) are shown in the purple bars in Fig. 11c, the column average dust concentration difference over the TP between the strong DOD years and the low DOD years is 5.42 μg m−3, which reaches 84% of the dust concentration during the low DOD years. The contributions of different sources to the column average dust concentration difference are shown as orange bars in Fig. 11c, which implies that the variation in the dust over the TP is attributable to the dust originating from EA and NA, accounting for 58% and 35% of the total variation, respectively.
c. Related mechanisms of the dust variation over the TP
To further analyze the reasons for the impact of dust from EA and NA on the dust variation over the TP, differences of wind vectors, East Asian dust concentrations, and North African dust concentrations between the SDY and the LDY are shown in Fig. 12. According to Figs. 12a1–a3, an abnormal cyclone appears in the northeast Atlantic, which causes increased southwesterly winds over NA and increased westerlies over the middle and high latitudes, resulting in increased transport of North African dust to the TP. Furthermore, the strengthening of northerly winds over the northern slope of the TP promotes the local emission over the TP and the transport of East Asian dust to the TP (Figs. 12b1–b3), which also promotes the movement of North African dust through the northern slope into the TP (Figs. 12a2,a3).
(a1)–(a3) Distribution of the anomaly fields of wind vectors (vectors; m s−1) and North African dust concentration (shading; μg m−3) between the SDY and the LDY at 600, 400, and 250 hPa. (b1)–(b3) As in (a1)–(a3), but for East Asian dust concentration.
Citation: Journal of Climate 37, 10; 10.1175/JCLI-D-23-0486.1
Regression analyses between the time series of DOD anomaly over the TP and geopotential heights as well as wind vectors during the springs of 2000–20 indicate the abnormal cyclone over the northeast Atlantic and the abnormal cyclone over Siberia enhance the westerlies from NA to the TP and northerly winds over the northern slope of the TP (Fig. 13a). Furthermore, the SST over the northern North Atlantic is significantly and positively correlated with the time series of DOD anomaly over the TP during the springs of 2000–20, and according to the time series of SST and DOD anomalies, there is a consistent change in the northern North Atlantic SST and DOD over the TP (Fig. 13b). Based on the time series of SST anomaly over the northern North Atlantic (the orange line in Fig. 13b), the years with the top 25% highest SST are extracted, which include 2007, 2008, 2010, 2006, and 2005. The composite differences in the Takaya and Nakamura (T-N) wave activity fluxes (Fig. 13c) and dust column mass density as well as dust column mass flux (Fig. 13d) between the field during selected positive SST years and the average field during 2000–20 indicates that with high SST over the northern North Atlantic, the strong Rossby wave propagates eastward from the northeast Atlantic to NA, causing a cyclone over the northeast Atlantic (Fig. 13a), then transports eastward to the northern side of the TP, and finally southward into the TP (Fig. 13c). According to the corresponding dust column mass density and mass flux anomalies, an abnormally large amount of dust originates over NA and transports eastward, subsequently reaching the northern side of the TP across Eurasia, and finally enters the TP with dust originating from other sources along the way based on the northwesterly winds. Besides, as shown in Fig. 13e, during the increased dust concentrations years, the updrafts over the northern slope of the TP are more intense, which can be attributed to stronger northerly winds causing stronger uplift blocked by the high topography of the northern slope of the TP. Enhanced northerly winds and updrafts intensify the emission and transport of East Asian dust, including the transport of external dust to the TP, as well as local emissions from the TP. Regression analyses also indicate the dust content over the TP is significantly and positively correlated with the vertical velocity over the northern slope of the TP (Fig. 13f). The SST over the northern North Atlantic and the vertical velocity over the northern slope of the TP can statistically explain 72% of the variances in DOD over the TP during the springs of 2000 to 2020, according to multiple linear regression.
(a) The regression patterns of the geopotential height (shading; m) and the wind vectors (vectors; m s−1) onto the time series of DOD anomaly over the TP (Fig. 3e) at 500 hPa in the spring during 2000–20. (b) As in (a), but for the SST (°C); the red rectangle represents the northern North Atlantic, the line graph indicates time series of standardized DOD anomalies over the TP (the purple line), and the time series of standardized northern North Atlantic SST anomalies (the orange line) during the spring from 2000 to 2020. The linear correlation coefficient is indicated by R, which is significant at the 95% confidence level. (c) Distributions of the anomaly field of the T-N wave activity flux (vectors; m2 s−2) and corresponding streamfunction anomalies (shading; 106 m2 s−1) with increased spring SST over the northern North Atlantic, relative to the average field from 2000 to 2020. (d) As in (c), but for dust column mass density (shading; g kg−1) and dust column mass flux (shading; g m−1 s−1). (e) Distribution of the anomaly fields of the simulated average vertical velocity (cm s−1) at lower layers (0–2 km above ground) between the SDY and the LDY. (f) The regression pattern of the average vertical velocity (cm s−1) onto the time series of DOD anomaly over the TP. The dots represent the regression coefficients are significant above the 95% confidence level.
Citation: Journal of Climate 37, 10; 10.1175/JCLI-D-23-0486.1
4. Conclusions and discussion
Using multisource data and a numerical model, this study analyzed the impact of five main dust sources on the distribution of and variation in the dust over the TP during the spring of 2000–20. By individually simulating the dust aerosols from each dust source during nine years, the contributions of different sources to the dust over the TP are detected, which indicates that dust originating from EA, ME, and NA are the main contributors to the spring dust over the TP, accounting for about 42%–68%, 13%–27%, and 9%–25% of the total dust concentration, respectively. Some previous studies based on numerical models have provided the proportion of dust from different sources in the dust concentration over the TP (Mao et al. 2019; Hu et al. 2020). Although there are differences in values of proportion, as Mao et al. (2019) reported the East Asian source and the North African source contributed 60% and 22% of the total dust over the TP in spring during 2010–14, respectively, nevertheless both the previous studies and our results indicate that East Asian dust has the greatest contribution to the dust over the TP, and also indicate that the dust over the TP is affected by multiple dust sources. Moreover, in our study, the contributions of five dust sources were analyzed, and our results are based on a 9-yr simulation, longer than the simulation time in previous studies. In addition, the selected 9 years include 3 years with strong DOD, 3 years with weak DOD, and 3 years with average DOD, so our study is more representative.
East Asian dust including local emissions over the TP and external transport enters the TP mainly with northerly winds and updrafts over the northern slope of the TP, making a substantial contribution to the dust over the northern part of the TP. The westerlies over midlatitudes and updrafts over the southwestern slope of the TP are the main entry points for dust from the ME and NA which significantly contributes to the dust over the southern and western parts of the TP. Additionally, as the deserts of ME and NA are located upstream of and far from the TP, dust from these sources can be advected into the TP after reaching the middle and upper layers, which results in their greatest contributions to dust over the TP in the upper layers (above 300 hPa). In contrast, as the East Asian deserts are close to the TP, the dust from EA at upper layers is more transported to downstream of the TP under the influence of westerlies, causing a decrease in the contribution of East Asian dust to the dust over the TP with increasing altitude.
Since the study periods were selected based on different DOD intensities, we in turn obtained the contributions of different sources to the variation in dust over the TP. East Asian and North African dust are the dominant factors, with potential contributions of 58% and 35% of the total variation, respectively. According to further analysis, the high SST over the northern North Atlantic and strong updrafts over the northern slope of the TP are primarily responsible for the variation in dust over the TP caused by dust from EA and NA. The former enhances the westerlies from NA to EA and the northwesterly winds over the northern slope of the TP and combines with the latter to promote the transport of dust from NA and EA to the TP. The main conclusions of this study are proposed in Fig. 14.
(a) Schematic illustration of the dust transport from different sources to the TP. The yellow arrows represent the approximate transport processes. The yellow numbers indicate the proportion of different dust sources in the dust concentration over the TP. (b) Schematic diagram illustrating the processes responsible for the influence of northern North Atlantic SST on the dust variation over the TP.
Citation: Journal of Climate 37, 10; 10.1175/JCLI-D-23-0486.1
Our study provides some evidence that against the backdrop of climate change, the dust over the TP is primarily affected by East Asian and North African dust. It should be noted that due to the limitations of numerical models, there may be some uncertainties in the results. As reported by Kok (2011), due to the overestimation of fine dust (diameter < 2 μm) emissions, existing numerical models may underestimate the total dust emissions, which will lead to an underestimation of the dust transport from sources close to the TP to the plateau. In addition, because the overestimated fine particles have the longest lifetime, the contribution of dust sources far away from the TP to the dust concentration over the plateau may be overestimated. Therefore, the error in dust emissions will affect our results. Furthermore, to reduce uncertainties based on simulations and further explain the mechanism, it is necessary to conduct large-scale dust observations from NA to EA.
Acknowledgments.
This research was supported by the Shandong Provincial Natural Science Foundation (ZR2023QD169), the Fundamental Research Funds for the Central Universities (202313022), the Second Tibetan Plateau Scientific Expedition and Research Program (2019QZKK0602), and the National Natural Science Foundation of China (41991231) as well as the Fundamental Research Funds for the Central Universities (lzujbky-2020-kb02).
Data availability statement.
The MISR observations used in this study were obtained from NASA (https://l0dup05.larc.nasa.gov/L3Web/download). The ERA5 reanalysis data can be gotten from the Climate Data Store (https://cds.climate.copernicus.eu/cdsapp#!/dataset/reanalysis-era5-pressure-levels-monthly-means). The MERRA-2 reanalysis data were obtained from NASA (https://goldsmr5.gesdisc.eosdis.nasa.gov/data/MERRA2/M2I3NVAER.5.12.4/). SST data were downloaded from the Hadley Centre (https://www.metoffice.gov.uk/hadobs/hadisst/data/download.html). The code of NICAM used in this study can be gotten from https://nicam.jp/dokuwiki/doku.php?id=Top. FNL data were obtained from NCEP (https://rda.ucar.edu/datasets/ds083.2/).
REFERENCES
Che, H., and Coauthors, 2015: Ground-based aerosol climatology of China: Aerosol optical depths from the China Aerosol Remote Sensing Network (CARSNET) 2002–2013. Atmos. Chem. Phys., 15, 7619–7652, https://doi.org/10.5194/acp-15-7619-2015.
Chen, S., J. Huang, C. Zhao, Y. Qian, L. R. Leung, and B. Yang, 2013: Modeling the transport and radiative forcing of Taklimakan dust over the Tibetan Plateau: A case study in the summer of 2006. J. Geophys. Res. Atmos., 118, 797–812, https://doi.org/10.1002/jgrd.50122.
Dai, T., G. Shi, and T. Nakajima, 2015: Analysis and evaluation of the global aerosol optical properties simulated by an online aerosol-coupled non-hydrostatic icosahedral atmospheric model. Adv. Atmos. Sci., 32, 743–758, https://doi.org/10.1007/s00376-014-4098-z.
Dai, T., Y. Cheng, P. Zhang, G. Shi, M. Sekiguchi, K. Suzuki, D. Goto, and T. Nakajima, 2018: Impacts of meteorological nudging on the global dust cycle simulated by NICAM coupled with an aerosol model. Atmos. Environ., 190, 99–115, https://doi.org/10.1016/j.atmosenv.2018.07.016.
Dong, Z., Z. Li, F. Wang, and M. Zhang, 2009: Characteristics of atmospheric dust deposition in snow on the glaciers of the eastern Tien Shan, China. J. Glaciol., 55, 797–804, https://doi.org/10.3189/002214309790152393.
Dong, Z., D. Qin, S. Kang, J. Ren, J. Chen, X. Cui, Z. Du, and X. Qin, 2014: Physicochemical characteristics and sources of atmospheric dust deposition in snow packs on the glaciers of western Qilian Mountains, China. Tellus, 66B, 20956, https://doi.org/10.3402/tellusb.v66.20956.
Dong, Z., J. Brahney, S. Kang, J. Elser, T. Wei, X. Jiao, and Y. Shao, 2020: Aeolian dust transport, cycle and influences in high-elevation cryosphere of the Tibetan Plateau region: New evidences from alpine snow and ice. Earth-Sci. Rev., 211, 103408, https://doi.org/10.1016/j.earscirev.2020.103408.
Feng, X., and Coauthors, 2020: Increased dust aerosols in the high troposphere over the Tibetan Plateau from 1990s to 2000s. J. Geophys. Res. Atmos., 124, e2020JD032807, https://doi.org/10.1029/2020JD032807.
Gelaro, R., and Coauthors, 2017: The Modern-Era Retrospective Analysis for Research and Applications, version 2 (MERRA-2). J. Climate, 30, 5419–5454, https://doi.org/10.1175/JCLI-D-16-0758.1.
Han, Y., and Coauthors, 2022: CALIOP-based quantification of Central Asian dust transport. Remote Sens., 14, 1416, https://doi.org/10.3390/rs14061416.
Hersbach, H., and Coauthors, 2020: The ERA5 global reanalysis. Quart. J. Roy. Meteor. Soc., 146, 1999–2049, https://doi.org/10.1002/qj.3803.
Hu, Z., J. Huang, Z. Zhao, Q. Jin, Y. Ma, and B. Yang, 2020: Modeling dust sources, transport, and radiative effects at different altitudes over the Tibetan Plateau. Atmos. Chem. Phys., 20, 1507–1529, https://doi.org/10.5194/acp-20-1507-2020.
Huang, J., and Coauthors, 2007: Summer dust aerosols detected from CALIPSO over the Tibetan Plateau. Geophys. Res. Lett., 34, L18805, https://doi.org/10.1029/2007GL029938.
Huang, J., P. Minnis, B. Chen, Z. Huang, Z. Liu, Q. Zhao, Y. Yi, and J. K. Ayers, 2008: Long-range transport and vertical structure of Asian dust from CALIPSO and surface measurements during PACDEX. J. Geophys. Res., 113, D23212, https://doi.org/10.1029/2008JD010620.
Jia, R., Y. Liu, B. Chen, Z. Zhang, and J. Huang, 2015: Source and transportation of summer dust over the Tibetan Plateau. Atmos. Environ., 123, 210–219, https://doi.org/10.1016/j.atmosenv.2015.10.038.
Kang, S., Y. Zhang, Y. Zhang, B. Grigholm, S. Kaspari, D. Qin, J. Ren, and P. Mayewski, 2010: Variability of atmospheric dust loading over the central Tibetan Plateau based on ice core glaciochemistry. Atmos. Environ., 44, 2980–2989, https://doi.org/10.1016/j.atmosenv.2010.05.014.
Kok, J. F., 2011: A scaling theory for the size distribution of emitted dust aerosols suggests climate models underestimate the size of the global dust cycle. Proc. Natl. Acad. Sci. USA, 108, 1016–1021, https://doi.org/10.1073/pnas.1014798108.
Lau, K.-M., and K.-M. Kim, 2006: Observational relationships between aerosol and Asian monsoon rainfall, and circulation. Geophys. Res. Lett., 33, L21810, https://doi.org/10.1029/2006GL027546.
Lau, W. K. M., C. Yuan, and Z. Li, 2018: Origin, maintenance and variability of the Asian Tropopause Aerosol Layer (ATAL): The roles of monsoon dynamics. Sci. Rep., 8, 3960, https://doi.org/10.1038/s41598-018-22267-z.
Li, F., X. Wan, H. Wang, Y. J. Orsolini, Z. Cong, Y. Gao, and S. Kang, 2020: Arctic sea-ice loss intensifies aerosol transport to the Tibetan Plateau. Nat. Climate Change, 10, 1037–1044, https://doi.org/10.1038/s41558-020-0881-2.
Li, P., G. Wu, X. Zhang, N. Yan, and X. Zhang, 2019: Variation in atmospheric dust since 1950 from an ice core in the central Tibetan Plateau and its relationship to atmospheric circulation. Atmos. Res., 220, 10–19, https://doi.org/10.1016/j.atmosres.2018.12.030.
Liu, G., J. Li, and T. Ying, 2023: The shift of decadal trend in Middle East dust activities attributed to North Tropical Atlantic variability. Sci. Bull., 68, 1439–1446, https://doi.org/10.1016/j.scib.2023.05.031.
Liu, L., and Coauthors, 2019: Contrasting influence of Gobi and Taklimakan deserts on the dust aerosols in western North America. Geophys. Res. Lett., 46, 9064–9071, https://doi.org/10.1029/2019GL083508.
Liu, Q., Z. Huang, Z. Hu, Q. Dong, and S. Li, 2022: Long-range transport and evolution of Saharan dust over East Asia from 2007 to 2020. J. Geophys. Res. Atmos., 127, e2022JD036974, https://doi.org/10.1029/2022JD036974.
Liu, Y., Q. Zhu, J. Huang, S. Hua, and R. Jia, 2019: Impact of dust-polluted convective clouds over the Tibetan Plateau on downstream precipitation. Atmos. Environ., 209, 67–77, https://doi.org/10.1016/j.atmosenv.2019.04.001.
Liu, Y., J. Huang, T. Wang, J. M. Li, H. R. Yan, and Y. L. He, 2022: Aerosol-cloud interactions over the Tibetan Plateau: An overview. Earth-Sci. Rev., 234, 104216, https://doi.org/10.1016/j.earscirev.2022.104216.
Mao, R., Z. Hu, C. Zhao, D.-Y. Gong, D. Guo, and G. Wu, 2019: The source contributions to the dust over the Tibetan Plateau: A modeling analysis. Atmos. Environ., 214, 116859, https://doi.org/10.1016/j.atmosenv.2019.116859.
Martonchik, J. V., D. J. Diner, K. A. Crean, and M. A. Bull, 2002: Regional aerosol retrieval results from MISR. IEEE Trans. Geosci. Remote Sens., 40, 1520–1531, https://doi.org/10.1109/TGRS.2002.801142.
Osada, K., H. Iida, M. Kido, K. Matsunaga, and Y. Iwasaka, 2004: Mineral dust layers in snow at Mount Tateyama, central Japan: Formation processes and characteristics. Tellus, 56B, 382–392, https://doi.org/10.3402/tellusb.v56i4.16436.
Randles, C. A., and Coauthors, 2017: The MERRA-2 aerosol reanalysis, 1980 onward. Part I: System description and data assimilation evaluation. J. Climate, 30, 6823–6850, https://doi.org/10.1175/JCLI-D-16-0609.1.
Rayner, N. A., D. E. Parker, E. B. Horton, C. K. Folland, L. V. Alexander, D. P. Rowell, E. C. Kent, and A. Kaplan, 2003: Global analyses of sea surface temperature, sea ice, and night marine air temperature since the late nineteenth century. J. Geophys. Res., 108, 4407, https://doi.org/10.1029/2002JD002670.
Satoh, M., and Coauthors, 2014: The Non-Hydrostatic Icosahedral Atmospheric Model: Description and development. Prog. Earth Planet. Sci., 1, 18, https://doi.org/10.1186/s40645-014-0018-1.
Sun, E., and Coauthors, 2019a: Variation in MERRA-2 aerosol optical depth over the Yangtze River delta from 1980 to 2016. Theor. Appl. Climatol., 136, 363–375, https://doi.org/10.1007/s00704-018-2490-9.
Sun, E., X. Xu, H. Che, Z. Tang, K. Gui, L. An, C. Lu, and G. Shi, 2019b: Variation in MERRA-2 aerosol optical depth and absorption aerosol optical depth over China from 1980 to 2017. J. Atmos. Sol.-Terr. Phys., 186, 8–19, https://doi.org/10.1016/j.jastp.2019.01.019.
Suzuki, K., T. Nakajima, M. Satoh, H. Tomita, T. Takemura, T. Y. Nakajima, and G. L. Stephens, 2008: Global cloud-system-resolving simulation of aerosol effect on warm clouds. Geophys. Res. Lett., 35, L19817, https://doi.org/10.1029/2008GL035449.
Takaya, K., and H. Nakamura, 2001: A formulation of a phase-independent wave-activity flux for stationary and migratory quasigeostrophic eddies on a zonally varying basic flow. J. Atmos. Sci., 58, 608–627, https://doi.org/10.1175/1520-0469(2001)058<0608:AFOAPI>2.0.CO;2.
Takemura, T., H. Okamoto, Y. Maruyama, A. Numaguti, A. Higurashi, and T. Nakajima, 2000: Global three-dimensional simulation of aerosol optical thickness distribution of various origins. J. Geophys. Res., 105, 17 853–17 873, https://doi.org/10.1029/2000JD900265.
Tomita, H., and M. Satoh, 2004: A new dynamical framework of nohydrostatic global model using the icosahedral grid. Fluid Dyn. Res., 34, 357, https://doi.org/10.1016/j.fluiddyn.2004.03.003.
Wang, M., W. K. M. Lau, and J. Wang, 2021: Impact of Middle East dust on subseasonal-to-seasonal variability of the Asian summer monsoon. Climate Dyn., 57, 37–54, https://doi.org/10.1007/s00382-021-05694-z.
Wang, T., and Coauthors, 2021: Identifying a transport mechanism of dust aerosols over South Asia to the Tibetan Plateau: A case study. Sci. Total Environ., 758, 143714, https://doi.org/10.1016/j.scitotenv.2020.143714.
Wu, G., T. Yao, B. Xu, L. Tian, C. Zhang, and X. Zhang, 2010: Dust concentration and flux in ice cores from the Tibetan Plateau over the past few decades. Tellus, 62B, 197–206, https://doi.org/10.1111/j.1600-0889.2010.00457.x.
Xu, C., Y. Ma, J. Ma, C. You, and H. Wang, 2020: Spring dust mass flux over the Tibetan Plateau during 2007–19 and connections with North Atlantic SST variability. J. Climate, 33, 9691–9703, https://doi.org/10.1175/JCLI-D-19-0481.1.
Xu, J., G. Yu, S. Kang, S. Hou, Q. Zhang, J. Ren, and D. Qin, 2012: Sr-Nd isotope evidence for modern aeolian dust sources in mountain glaciers of western China. J. Glaciol., 58, 859–865, https://doi.org/10.3189/2012JoG12J006.
Xu, X., T. Zhao, C. Lu, Y. Guo, B. Chen, R. Liu, Y. Li, and X. Shi, 2014: An important mechanism sustaining the atmospheric “water tower” over the Tibetan Plateau. Atmos. Chem. Phys., 14, 11 287–11 295, https://doi.org/10.5194/acp-14-11287-2014.
Xu, X., H. Wu, X. Yang, and L. Xie, 2020: Distribution and transport characteristics of dust aerosol over Tibetan Plateau and Taklimakan Desert in China using MERRA-2 and CALIPSO data. Atmos. Environ., 237, 117670, https://doi.org/10.1016/j.atmosenv.2020.117670.
Yao, T. D., and Coauthors, 2019: Recent Third Pole’s rapid warming accompanies cryospheric melt and water cycle intensification and interactions between monsoon and environment: Multidisciplinary approach with observations, modeling, and analysis. Bull. Amer. Meteor. Soc., 100, 423–444, https://doi.org/10.1175/BAMS-D-17-0057.1.
Zhang, H., X. Fu, B. Yu, B. Li, P. Liu, G. Zhang, L. Zhang, and X. Feng, 2021: Speciated atmospheric mercury at the Waliguan Global Atmosphere Watch station in the northeastern Tibetan Plateau: Implication of dust-related sources for particulate bound mercury. Atmos. Chem. Phys., 21, 15 847–15 859, https://doi.org/10.5194/acp-21-15847-2021.
Zhu, J., and Coauthors, 2019: Spatiotemporal variation of aerosol and potential long-range transport impact over the Tibetan Plateau, China. Atmos. Chem. Phys., 19, 14 637–14 656, https://doi.org/10.5194/acp-19-14637-2019.
Zhu, Q., Y. Liu, T. Shao, R. Luo, and Z. Tan, 2021: A simulation study on the new transport pathways of global tropopause dust layer. Geophys. Res. Lett., 48, e2021GL096063, https://doi.org/10.1029/2021GL096063.
