Ground-based vertical profile observations of atmospheric composition on the Tibetan Plateau (2017–2019)

The Tibetan Plateau (TP) plays an essential role in modulating regional and global climate, and its influence on climate is also affected by human-related processes, including changes in atmospheric composition. However, observations of atmospheric composition, especially vertical profile observations, remain sparse and rare on the TP, due to extremely high altitude, topographical heterogeneity and the grinding environment. Accordingly, the forcing and feedback of atmospheric composition from rapidly changing surrounding regions to regional environmental and climate change in the TP remains poorly understood. This paper introduces a high-time-resolution (∼15 min) vertical profile observational dataset of atmospheric composition (aerosols, NO₂, HCHO and HONO) on the TP for more than 1 year (2017–2019) using a passive remote sensing technique. The diurnal pattern, vertical distribution and seasonal variations of these pollutants are documented here in detail. The sharing of this dataset would benefit the scientific community in exploring source–receptor relationships and the forcing and feedback of atmospheric composition on the TP to the regional and global climate. It also provides potential to improve satellite retrievals and to facilitate the development and improvement of models in cold regions. The dataset is freely available at Zenodo (https://doi.org/10.5281/zenodo.5336460; Xing, 2021).

1 Introduction

The Tibetan Plateau (TP) and the surrounding Hindu Kush Himalayan mountains are referred to as the world's “Third Pole” due to their vast wealth of areas and glaciers (Qiu, 2008). The TP is the highest and largest plateau in the world with a ∼4000 m average elevation. Headwaters of six major rivers in Asia start on the TP (Kang et al., 2010; Cheng and Jin, 2013), and it thus acts the role of the “Water Tower of Asia”, determining the survival and development of more than 2 billion people (Xu et al., 2008; Immerzeel et al., 2010; J. Gao et al., 2019; Kang et al., 2019). A number of studies have suggested that the TP is an important driving force to thermally and dynamically affect global and regional climate (Yanai et al., 1992; Liu et al., 2007; Boos and Kuang, 2010). In contrast to the surrounding cold middle troposphere, the TP receives strong solar radiation at the surface owing to its topographic characteristics, leading to its critical role in the evolution and variability of South Asian and East Asian monsoons (Wu et al., 2007, 2012). It also acts as a natural barrier that inhibits northward moisture transport (Dong et al., 2017) and splits subtropical westerlies (Bolin, 1950). The TP is usually regarded as one of the most sensitive areas responding to global environmental and climate change (Peng et al., 2012) and can amplify the changes to a global scale (Kang et al., 2010).

Recent decades have witnessed increased warming, together with the retreat of glaciers and early melting of snow, on the TP (He et al., 2003; Lau et al., 2010; Xu et al., 2016), leading to serious side issues, including soil erosion, floods, etc. (Shrestha et al., 1999). As the warming rate of the TP is much faster than the greenhouse warming rate, it has been argued that other factors besides greenhouse warming might have resulted in accelerated warming over the TP (Kang et al., 2000; Ramanathan et al., 2007). The TP was usually considered a background region of atmospheric composition due to low population density and less intense human activities. However, studies in the past decade have suggested that transport of pollutants from South Asia, East Asia and Southeast Asia cannot be ignored (Cong et al., 2009; Yao et al., 2012; Zhang et al., 2015; Kang et al., 2019). Ample evidence has demonstrated the significance of atmospheric composition in influencing the retreat of glaciers on the TP (Xu et al., 2009), while large uncertainties remain in understanding the sources, mainly owing to lack of long-term observations that have been obstructed by complex topography and a harsh environment (Cong et al., 2015a; Barnett et al., 2005; Pu et al., 2007; Bolch et al., 2012; Kang et al., 2016).

To fill these gaps, a series of field campaigns have been carried out in recent years to unveil the features of weather and atmospheric composition over the TP, especially after the “second comprehensive scientific expedition to the Qinghai–Tibet Plateau” plan (Wang et al., 2021; Che and Zhao, 2021; H. Liu et al., 2021). The Institute of Tibetan Plateau Research, Chinese Academy of Sciences (ITP, CAS), established six long-term stations to measure both meteorological and micrometeorological variables over the TP (Ma et al., 2020a, b). The China National Environmental Monitoring Center (CNEMC) has also established an atmospheric composition monitoring network (Gao et al., 2020b; Liang et al., 2017) on the TP that has continuously monitored the ground-level concentrations of PM_2.5, PM₁₀, NO₂, SO₂, O₃ and CO since 2013, covering Lhasa, Shigatse, Qamdo, Nyingchi, Shannan, Nagqu and Ngari. Satellite retrievals (i.e., OMI, MODIS and CALIPSO) have also been widely applied to elucidate the spatial and temporal variation in atmospheric processes over the TP (Zhu et al., 2019; Li et al., 2020). Although the vertical profiles of aerosols were revealed with CALIPSO (Huang et al., 2007), the relatively long revisiting time (∼16 d) hinders understanding of processes at smaller scales. Accordingly, the current atmospheric composition monitoring platforms and freely available datasets are inadequate to fully understand the sources and impacts of atmospheric composition on the TP, as the formation, aging and transport processes also occur above the ground (Hindman and Upadhyay, 2002; Duo et al., 2018; Huang et al., 2007). If we use only satellite column or ground-level observations to infer transport fluxes of atmospheric trace gases, the relative significance of local and transport contributions could be overvalued/downplayed (Hu et al., 2020; Liu et al., 2021a). In addition, the low air density, land surface heterogeneity, relatively cold temperature and strong solar radiation on the TP cause a relatively high planetary boundary layer (PBL) compared to over its surrounding lowlands (Yang et al., 2003; Seidel et al., 2010). The conditions promote the exchange of atmosphere between the PBL and stratosphere, which has important impacts on atmospheric chemistry (Škerlak et al., 2014). The lack of sufficient vertical profile observation data limits the understanding of atmospheric composition on the TP and its impact on the global climate.

Limited field experiments have been conducted using balloon and lidar on the TP in recent years (Wu et al., 2016; Dai et al., 2018; Fang et al., 2019; Zhang et al., 2020). The instruments used are costly, and very limited atmospheric composition species (mostly O₃) can be measured. Additionally, these observations are not open for sharing or have very limited availability upon request (i.e., only data during specified periods are provided). The coarse temporal resolution of these data is not sufficient for evaluations of chemical transport modeling and climate modeling (Gao et al., 2020a). Multi-axis differential optical absorption spectroscopy (MAX-DOAS) uses scattered sunlight as the signal source, enabling it to achieve low-cost and continuous measurement of vertical profiles of atmospheric composition.

Here we describe and provide access to a high-time-resolution dataset of vertical profiles of atmospheric composition over the TP for more than 1 year. This database will play a critical role in improving satellite retrievals and numerical modeling of atmospheric composition over the TP. The uncertainties in the air mass factor (AMF) caused by the high surface albedo and the absence of an a priori profile on the TP could lead to large uncertainties in satellite retrievals of trace gases. Previously, we have demonstrated that using the AMF calculated with MAX-DOAS-measured NO₂ vertical profiles could remarkably improve the accuracy of retrievals of NO₂ column densities (VCDs) (Liu et al., 2016; Xing et al., 2017). On top of that, due to complex terrain and weather conditions, uncertainties in emission inventories, and imperfect model parameterization, chemical transport models face difficulties in capturing the vertical structures of atmospheric composition on the TP (Yang et al., 2018b), the uncertainties of which could be constrained through the assimilation of observations from this dataset. Section 2 describes the observation site, MAX-DOAS instrument and retrieval algorithms. The vertical profiles of aerosols, NO₂, HCHO and HONO and their diurnal and monthly variations are introduced in Sect. 3. Sections 4 and 5 present the availability of this dataset (Xing et al., 2021) and a summary, respectively.

2 Experimental setup

2.1 Description of the monitoring site

The MAX-DOAS instrument was set up at the Qomolangma Atmospheric and Environmental Observation and Research Station, CAS (QOMS) (4276 m a.s.l.; 28.21^∘ N, 86.56^∘ E), which is located at the bottom of the lower Rongbuk valley (Fig. 1). The geomorphic setting of this station is mainly alpine covered by sandy soil, and sparse vegetation exists. During the summer monsoon season, abundant moisture from the Indian Ocean prevails at CAS (QOMS), increasing humidity and precipitation there. The maximum humidity was observed from July to August, and lower values usually appeared from December to February (Zhang et al., 2018; Chen et al., 2018). To the south of the measurement site is the South Asian subcontinent, covering Nepal, India and Bangladesh. The South Asian subcontinent is currently one of the most polluted regions in the world. Averaged spatial distributions of satellite-monitored aerosol optical depth (AOD), NO₂ and HCHO from December 2017 to March 2019 are displayed in Fig. S1 in the Supplement. Elevated AOD, NO₂ and HCHO values are found in the southern foothills of the Himalayas compared to those on the Tibetan Plateau. Local circulations, such as valley winds, closely connect the near-surface atmosphere with the free atmosphere on the north side of Mt. Everest, and the South Asia summer monsoon could drive pollutants from the South Asian subcontinent to influence atmospheric composition over the TP. Accordingly, the CAS (QOMS) site acts as a terrific location to monitor the vertical distribution of atmospheric composition and to explore the sources and impacts of these pollutants.

Figure 1 Elevation map of the TP and location of the CAS (QOMS) monitoring site.

2.2 Description of the MAX-DOAS instrument

The MAX-DOAS instrument at CAS (QOMS) was operated from December 2017 to March 2019. It consisted of two key parts, namely the telescope scanner unit and the spectrometer unit (Fig. 2). Temperature stabilization of both spectrometers; USB communication and control electronics for additional sensors to record temperature, pressure and inclination; and the telescope elevation stepper motor were also incorporated.

Figure 2 (a) Physical view and (b) schematic view of the MAX-DOAS instrument.

The telescope unit contained motor, electronics, a MEMS inclination sensor and optical components. A quartz-glass tube was used to protect the rotating prism, which reflected the light through a lens onto the fiber at the end of an adjustable tube. This unit was controlled by a stepping motor to collect the scattered sunlight at different elevation angles, which was then transmitted via a glass fiber bundle to the spectrometer unit. The stepping motor had a precision of 0.01^∘, and the field of view (FOV) of the telescope was less than 0.3^∘. The telescope scanner unit covered elevation angles from −10 to 180^∘, with an angle accuracy of <0.1^∘. Moreover, an azimuth motor covering 0–180^∘ was installed in the telescope scanner unit to observe atmospheric composition at different azimuth angles. In this study, we set the elevation angle sequence to 1, 2, 3, 4, 5, 6, 8, 10, 15, 30 and 90^∘ and exposure time of each individual spectrum to 1 min. The telescope unit was pointed to an azimuth angle of 53^∘.

Figure 3 Examples of DOAS fits for O₄, NO₂, HCHO and HONO (∼ 13:30 LST on 22 May 2018; SZA and elevation angle (EVA) values were 52.53 and 15^∘, respectively).

The spectrometer unit included two AvaSpec-ULS2048L spectrometers (UV 300–460 nm, visible 460–630 nm) with a 0.6 nm spectral resolution. A cooling unit was used to control the spectrometer temperature. Typical instrumental stray light was <0.05 %, with a root mean square (rms) of $\mathrm{1}\times {\mathrm{10}}^{-\mathrm{4}}$ (visible) and $\mathrm{2}\times {\mathrm{10}}^{-\mathrm{4}}$ (UV) for ∼1000 scans around noon (11:00–13:00, UTC+8). A Peltier element was mounted on the spectrometer housing to cool and heat the spectrometer, and the temperature was stabilized at 20 ^∘C with low fluctuations of <0.05 ^∘C. We also equipped a charge-coupled device camera (Sony ILX511 with 2048 pixels) to the spectrometer to convert the analog signal to a digital signal. Moreover, the dark current and electric offset were recorded at night to correct the observed spectra. To avoid the strong absorption of the stratosphere, this study analyzed only spectra collected when the solar zenith angle (SZA) was less than 75^∘.

2.3 Spectral retrieval

This study used the QDOAS software developed by the Royal Belgian Institute for Space Aeronomy (BIRA-IASB) (http://uv-vis.aeronomie.be/software/QDOAS/, last access: 26 April 2021) to retrieve differential slant column densities (DSCDs) of the oxygen dimer (O₄), NO₂, HCHO and HONO. A sequential zenith spectrum, estimated from interpolation of two zenith spectra recorded before and after an elevation sequence, was selected as the Frauenhofer reference spectrum. The retrieval configuration settings followed Xing et al. (2020, 2021) and the configurations used in the CINDI intercomparison campaign as well (Roscoe et al., 2010; Kreher et al., 2020; Y. Wang et al., 2020) (details are listed in Table 1). Figure 3 illustrates the DOAS fits of the above four species, and reasonably good fitting with tiny values of the rms can be found. Retrieved data with rms values larger than $\mathrm{5}\times {\mathrm{10}}^{-\mathrm{4}}$, $\mathrm{5}\times {\mathrm{10}}^{-\mathrm{4}}$, $\mathrm{6}\times {\mathrm{10}}^{-\mathrm{4}}$ and $\mathrm{5}\times {\mathrm{10}}^{-\mathrm{3}}$ for O₄, NO₂, HCHO and HONO, respectively, were filtered out. We also calculated the color index (CI), defined as the ratio of spectral intensities at 330 nm to those at 390 nm, to remove the cloud effects (Wagner et al., 2016). We filtered out data when the CI was less than 10 % of the threshold that was obtained through fitting a fifth-order polynomial to CI data (Ryan et al., 2018). After these processes, 90.17 %, 86.41 %, 83.22 % and 80.19 %, respectively, of the original DSCDs data were marked as qualified.

Table 1 Detailed retrieval settings of O₄, NO₂, HCHO and HONO.

^* Solar I₀ correction, Aliwell et al. (2002).

2.4 Vertical profile retrieval algorithm

The vertical profiles of aerosol extinction and volume mixing ratios (VMRs) of trace gases (i.e., NO₂, HCHO and HONO) were retrieved using the optimal estimation method (OEM)-based algorithm. The radiative transfer model VLIDORT (Spurr, 2006) was used as the forward model. The maximum posteriori state vector x was determined through optimizing the following cost function χ².

$$\begin{array}{}\text{(1)}& {\mathit{\chi }}^{\mathrm{2}}=(\mathit{y}-\mathbf{F}\mathit{x},\mathit{b}){)}^T{\mathbf{S}}_{\mathit{\epsilon }}^{-\mathrm{1}}(\mathit{y}-\mathbf{F}(\mathit{x},\mathit{b}\left)\right)+{\left(\mathit{x}-{\mathit{x}}_{\mathrm{a}}\right)}^T{\mathbf{S}}_{\mathrm{a}}^{-\mathrm{1}}\left(\mathit{x}-{\mathit{x}}_{\mathrm{a}}\right),\end{array}$$

where F(x,b) denotes the measurement vector y (DSCDs measured at different elevation angles) as a function of the state vector x (pollutant profiles) and the true atmospheric meteorological parameters (profiles of temperature and pressure, albedo, and aerosol phase function). x_a stands for the a priori state vector. S_ε and S_a represent the covariance matrices of y and x_a, respectively. During the retrieval process, an exponentially decreasing shape was assumed as the initial a priori vertical profile shape of aerosols and trace gases, and the corresponding WRF-Chem-simulated AOD and VCDs were also used as input a priori information. The inversion strategy used a Gauss–Newton (GN) scheme (Wedderburn, 1974). The iterative weighting function K was calculated using the Jacobians of DSCDs. The inversion consisted of two steps. The aerosol vertical profile was retrieved first and then fed into the forward model to retrieve the trace gases' profile.

In this study, we derived the vertical profiles of aerosols and trace gases at 30 vertical layers, covering from 0.0 to 3.0 km, and a 1.0 km correlation length was chosen. For forward simulations, the surface albedo and surface altitude were set as 0.08 and 4.2 km, respectively. In addition, a fixed single-scattering albedo (SSA) of 0.85 and an aerosol phase function with an asymmetry parameter of 0.65 were selected, due to the low uncertainties in retrieved aerosols with a fixed SSA and asymmetry parameter (Irie et al., 2008). The profiles of aerosols and trace gases were filtered out when the degree of freedom (df) was less than 1.0 and the retrieved relative error larger than 100 %.

3 Vertical profile observations of atmospheric composition

Figure 4 illustrates the monthly data integrity of vertical profiles of aerosols, NO₂, HCHO and HONO at CAS (QOMS). The value of 100 indicates completely continuous observations. Better data integrity occurred mostly in autumn, and serious missingness happened mainly in summer. Missing data were largely associated with power outages of the instrument or the measurement station, technical problems of the instrument, poor data quality under unfavorable weather conditions (i.e., rain or high cloud coverage), and retrieval failure (i.e., high retrieval error or low df's). Considering the heterogeneity of atmospheric composition in the vertical direction, the resolutions of vertical profiles within 0–1.0 and 1.0–3.0 km were set to 0.1 and 0.2 km, respectively. During the day, the PBL over the TP could reach as high as 1–3 km (Chen et al., 2016; Xu et al., 2019). We thus used three representative layers, namely 0–100, 500–600 and 900–1000 m, to represent the bottom, middle and upper boundary layers.

Figure 4 Monthly data integrity of the vertical profiles of aerosols, NO₂, HCHO and HONO.

3.1 Aerosols

As indicated in Fig. 5, the maximum AOD over the TP occurred in August (1.19), which was almost double the minimum that happened in April (0.57). However, the high levels of AOD in summer might have been overestimated due to the poor integrity of AOD data in summer (Fig. 4, 12.91 % in July and 22.58 % in August). Relatively enhanced levels of AOD started from September and persisted till February. Such an enrichment was likely to be associated with both the increased anthropogenic emissions due to tourism and the transported pollution plume from South Asia. The ground surface of CAS (QOMS) is exposed continually throughout the year and is rarely covered by ice, snow or vegetation. Valley and mountain winds in this region began to strengthen from October, and strong winds persisted throughout the whole winter to blow soil and dust particles into the atmosphere (Xu et al., 2015; Liu et al., 2017; Chen et al., 2018).

Figure 5 Time series of monthly AOD from December 2017 to February 2019 at CAS (QOMS).

Figure 6a shows the time series of the hourly averaged aerosol extinction coefficient at layers of 0–100, 500–600 and 900–1000 m from December 2017 to February 2019. It suggests that the elevated abundance of aerosols was mainly concentrated in the lower layer and aerosol loadings decreased with height. Consistent with the seasonality shown in Fig. 5, high aerosol extinction coefficients appeared frequently in autumn and winter. Figure 6a also indicates that strong aerosol extinction coefficients occurred occasionally in the middle layer, associated with the long-range transport of particles (Zhang et al., 2018; Chen et al., 2018). The total-averaged and seasonally averaged vertical profiles of aerosol extinction are displayed in Fig. 7. All averaged profiles exhibit an exponentially decreasing shape with maximum values occurring near the surface. The aerosol extinction coefficient in the lower layer varied with seasons, with a maximum in autumn (1.55 km⁻¹) and minimum in spring (1.00 km⁻¹). The ratios of aerosol extinction coefficients in the middle layer to those in the lower boundary layer were 44.96 %, 41.11 %, 44.35 %, 40.01 % and 52.31 % for the total-averaged, spring-averaged, summer-averaged, autumn-averaged and winter-averaged aerosol profiles, respectively. These numbers decreased to 24.81 %, 22.72 %, 25.87 %, 21.29 % and 30.02 % for the upper layer. Relatively high levels of aerosols above ground during winter are associated with strong mountain-valley breeze, which can blow dust on the ground surface into high layers (Liu et al., 2017). Aerosol pollution is also more severe in winter in South Asia (M. Gao et al., 2019) and can be transported as an important source of aerosols over the TP (Xia et al., 2011; Lu et al., 2012; Cong et al., 2015a, b; Li et al., 2016; Chen et al., 2018; Zhang et al., 2018).

Figure 6 Time series of hourly averaged (a) aerosol extinction, (b) NO₂, (c) HCHO and (d) HONO from December 2017 to February 2019 at CAS (QOMS).

Figure 7 (a) Total-averaged, (b) spring-averaged, (c) summer-averaged, (d) autumn-averaged and (e) winter-averaged aerosol extinction vertical profiles from December 2017 to February 2019 at CAS (QOMS).

Figure 8a–e illustrate the diurnal variations in aerosol vertical profiles for different seasons. Lower aerosol extinction mainly occurred before 10:00 BJT, and it gradually increased and spread to higher altitudes with the rise of the planetary boundary layer (PBL) height after 10:00 BJT. Moreover, aerosol extinction showed bi-peak patterns in all four seasons. One peak appeared between 10:30–12:30 BJT, and the other one occurred after 14:00 BJT. This bi-peak pattern was in line with previous investigation by Liu et al. (2017), dominated by the effects of local aerosol emissions, local dust geomorphology and mountain-valley breeze (Liu et al., 2017). We observed also that the diffusion height of aerosols exhibited a maximum in summer and minimum in winter, mainly due to differences in PBL height driven by temperature in four seasons.

Figure 8 Diurnal variations in the total-averaged and seasonally averaged aerosol extinction profiles (a–e), NO₂ profiles (f–j), HCHO profiles (k–o) and HONO profiles (p–t) from December 2017 to February 2019 at CAS (QOMS).

3.2 Nitrogen dioxide (NO₂)

Figure 6b presents the time series of hourly NO₂ concentration at 0–100, 500–600 and 900–1000 m layers from December 2017 to February 2019. Abundance of NO₂ in the lower, middle and upper boundary layers exhibited clearly different levels, and elevated levels were concentrated in the lower levels. Higher NO₂ concentrations were observed in May and September–December (Chen et al., 2019), which was likely associated with enhanced vehicle emissions due to the arrival of the tourist season in Tibet from May. Additionally, the transmission of NO₂ emitted from the Indo-Gangetic Plain (Chen et al., 2017; Yang et al., 2018a) and the enhanced local anthropogenic emissions (i.e., heating using cow dung as fuel) in cold periods also caused the relatively high NO₂ levels.

The vertical structure of NO₂ also presented an exponentially decreasing shape, with the highest concentration located in the lower layer (Fig. 9). NO₂ concentrations in the lower layer peaked in autumn (1.28 ppb), and a minimum happened in spring (0.82 ppb). The ratios of NO₂ concentrations in the middle layer to those in the lower boundary layer were 39.50 %, 38.14 %, 40.19 %, 40.16 % and 39.39 % for the total-averaged, spring-averaged, summer-averaged, autumn-averaged and winter-averaged aerosol profiles, respectively. These numbers declined to 18.10 %, 17.36 %, 18.50 %, 18.44 % and 18.07 % for the upper layer. Differently from aerosol extinction, higher levels of NO₂ above ground occurred mainly in summer, instead of winter, which might be associated with stronger mixing within the PBL in summer.

Figure 9 (a) Total-averaged, (b) spring-averaged, (c) summer-averaged, (d) autumn-averaged and (e) winter-averaged NO₂ vertical profiles from December 2017 to February 2019 at CAS (QOMS).

The diurnal variation in NO₂ in spring, summer and winter also exhibited a bi-peak pattern (Fig. 8f–j), with one peak within 11:00–13:00 BJT and the other after 16:00 BJT. These two periods are usually accompanied by a low PBL height. Lower NO₂ concentration in the morning (before 10:00 BJT) can be attributed to fewer vehicle emissions (Wang et al., 2011). The elevated PBL height was conducive to the diffusion and dilution of NO₂ and led to lower NO₂ concentrations from 13:00 to 16:00 BJT. In autumn, another two NO₂ peaks values were observed between 14:00–16:00 BJT, which were likely to be related to tourist traffic.

Figure 10 (a) Total-averaged, (b) spring-averaged, (c) summer-averaged, (d) autumn-averaged and (e) winter-averaged HCHO vertical profiles from December 2017 to February 2019 at CAS (QOMS).

3.3 Formaldehyde (HCHO)

Formaldehyde (HCHO) is one of the most abundant volatile organic compounds (VOCs) in the atmosphere, and it strongly correlates with peroxy radicals (Kleinman et al., 2001; Ling et al., 2017). HCHO plays a critical role in atmospheric photochemistry to strongly drive ozone (O₃) formation (Luecken et al., 2012; Hassan et al., 2018). It comes from both primary sources (i.e., fossil fuel combustion, biomass burning, traffic and industrial activities) and secondary production (i.e., the oxidation of biogenic VOCs). Figure 6c displays the time series of hourly averaged HCHO concentration at 0–100, 500–600 and 900–1000 m layers from December 2017 to February 2019. We found HCHO at CAS (QOMS) was mainly distributed in the lower boundary layer; its concentrations decreased with height, similarly to NO₂. The highest HCHO concentrations mainly appeared in September–November, and the lowest HCHO concentrations occurred in December–February. The highest amount of vegetation appeared in autumn at CAS (QOMS), and isoprene along with other active VOCs emitted from vegetation can accumulate near the surface and be oxidized to form high levels of HCHO (Mu et al., 2007). On the other hand, local outdoor biomass burning is active in this season (Chen et al., 2018). The incomplete combustion could emit methane and non-methane volatile organic compounds (NMVOCs) to enhance HCHO concentrations (Gonzi et al., 2011). Fewer primary and secondary sources of HCHO were observed in winter.

Figure 11 (a) Total-averaged, (b) spring-averaged, (c) summer-averaged, (d) autumn-averaged and (e) winter-averaged HONO vertical profiles from December 2017 to February 2019 at CAS (QOMS).

Figure 10 provides the total-averaged and seasonally averaged vertical profiles of HCHO. Similarly to NO₂, vertical profiles of HCHO in all seasons an displayed exponentially decreasing shape, with the highest concentration in the lower boundary layer (total-averaged value being 3.15 ppb). The HCHO concentrations in the lower layer peaked in autumn (5.20 ppb) and dropped in winter (2.01 ppb). However, the vertical gradient of HCHO was smaller than that of NO₂. The ratios of HCHO concentrations in the middle layer to those in the lower boundary layer were 62.86 %, 69.74 %, 68.06 %, 60.00 % and 60.20 % for the total-averaged, spring-averaged, summer-averaged, autumn-averaged and winter-averaged aerosol profiles, respectively. These numbers dropped to 42.86 %, 47.60 %, 46.27 %, 40.77 % and 41.14 % for the upper layer. The vertical gradient of HCHO in autumn was larger than that in the other three seasons, which was partially associated with stronger surface HCHO sources in autumn.

On the diurnal variations in vertical profiles of HCHO, lower concentrations were observed before 10:00 BJT, but the abundance gradually increased with sunrise and the elevation of temperature (Fig. 8k–o). HCHO peaked within 10:00–16:00 BJT in spring and winter, while peak values of HCHO also appeared after 16:00 BJT due to the prolonged daytime in summer and autumn. The maximum and minimum diffusion heights of HCHO appeared in summer and winter, respectively, associated with PBL height.

3.4 Nitrous acid (HONO)

The photolysis of nitrous acid (HONO) is a significant source of the hydroxyl (OH) radical. HONO mainly originates from primary emissions from vehicles, ships, biomass burning and soil, the photolysis of nitrate particles (${\mathrm{NO}}_{\mathrm{3}}^{-}$), the homogeneous reaction of NO with OH radical, and the heterogeneous reaction of NO₂ on several types of surfaces (Wang et al., 2015; Fu et al., 2019). As shown in Fig. 6d, elevated HONO was mostly distributed in the lower boundary layer, and higher HONO concentrations appeared from July 2018 to February 2019. Enhanced levels after summer were likely connected with elevated direct emissions from vehicles, biomass burning and soil (Wang et al., 2015; Fu et al., 2019). We find high aerosol extinction, NO₂ and HONO concentrations simultaneously in Fig. 8d, i and s, suggesting that the heterogeneous reaction of NO₂ on aerosol surfaces might be another important source of HONO over the TP.

Figure 11 shows that the vertical profiles of HONO in all seasons exhibited an exponential shape, with the highest concentration (0.93 ppb) in the lower boundary layer. HONO concentrations in the lower layer peaked in summer (1.11 ppb) and dropped in spring (0.56 ppb) (Fig. 11). Differently from aerosols, NO₂ and HCHO, HONO concentrations in the middle and upper layers were extremely low. The proportions of HONO concentrations in the middle layer to those in the lower boundary layer were 8.86 %, 13.68 %, 7.82 %, 9.05 % and 10.44 % for the total-averaged, spring-averaged, summer-averaged, autumn-averaged and winter-averaged aerosol profiles, respectively. These numbers decreased to 3.14 %, 5.30 %, 3.52 %, 2.72 % and 3.06 % for the upper layer. On the diurnal variations in vertical profiles of HONO, we found that HONO concentrations were, consistently in all four seasons, lower before 10:00 BJT and peaked around 10:00–14:00 BJT. This pattern in the monitoring site was different from that in other low-altitude cities (Wang et al., 2013, 2017; Lee et al., 2016; Xing et al., 2021).

Figure 12 Linear regression plots (a) of the tropospheric NO₂ VCDs measured by OMI and MAX-DOAS and (b) of the tropospheric HCHO VCDs measured by OMI and MAX-DOAS.

3.5 Validations against satellite retrievals

In this study, OMI NO₂ and HCHO products were used to validate the described dataset. For better comparison, MAX-DOAS data were averaged over the OMI overpass time (from 06:30 to 08:30 BJT) in the region that covers CAS (QOMS). OMI data were spatially averaged over the 15 km grid cells around the MAX-DOAS site, as the spatial resolution of OMI is 15 km × 15 km. The VCDs of NO₂ and HCHO were derived by vertical integrations of their vertical profiles. The linear association between MAX-DOAS and OMI measurements are shown in Fig. 12. We found reasonably good correlations between MAX-DOAS and OMI observations, with Pearson correlation coefficients (r) of 0.68 and 0.60 (regression coefficients (R) of 1.21 and 1.12, offset of 1.29×1015 and 2.06×1015 molec. cm⁻²) for NO₂ and HCHO, respectively. OMI NO₂ and HCHO VCDs were systematically lower than MAX-DOAS NO₂ and HCHO VCDs, which is different from previous comparisons applied in lower-altitude areas (Liu et al., 2016; Xing et al., 2017). This might be caused by the large-area averaging effect of OMI and uncertainties in OMI products introduced by cloud cover and high surface albedo.

We also observed that the correlations between MAX-DOAS and OMI observations exhibited significant differences across seasons. The highest correlation appeared in autumn, followed by winter (Table S1 in the Supplement). High correlation in these two seasons was associated with increased tourism vehicle emissions, regional transport of NO₂ and enhanced photochemical formation of HCHO. Failed linear regression analysis mainly appeared in spring and summer, due to high cloud coverage from March to September on the TP (Chen and Dong, 2006).

4 Data availability

Ground-based vertical profile observations of atmospheric composition on the Tibetan Plateau introduced here are available in .txt format at Zenodo (https://doi.org/10.5281/zenodo.5336460; Xing, 2021).

5 Summary

In situ observations, especially profile observations, are scarce but essential in polar regions, including the TP, the so-called Third Pole. The sources of atmospheric composition over the TP and how the species interact with regional and global climate remain unclear, mainly owing to lack of observations. Large uncertainties also remain in the chemical transport and climate modeling of cold-region processes, without observational constraints. The presented dataset offers high-time-resolution vertical profile observations of aerosols, NO₂, HCHO and HONO for more than 1 year on the TP. Distinct features were found in the diurnal variations, vertical distributions and seasonality of these atmospheric composition species. Data products were also validated with satellite retrievals.

The limitations of this data are described as follows. (1) MAX-DOAS in ultraviolet and visible spectral ranges are typically affected by photon shot noise, and the retrieval errors usually increase under heavy haze or cloudy conditions. The data with relative retrieval errors larger than 50 % were filtered out in this study. (2) Only the daytime vertical profiles of aerosols, NO₂, HCHO and HONO were retrieved since MAX-DOAS relies on scattered sunlight. The spectral data collected when solar zenith angles (SZAs) were larger than 75^∘ were filtered out to avoid the strong absorption of the stratosphere (Xing et al., 2017). (3) The vertical resolution of 100 m is the highest resolution at present but still needs to be improved with the development of hardware and algorithms in the future. This dataset provides better temporal coverage and fills the gaps in the vertical direction. It has potential to both improve satellite retrievals and reduce uncertainties in chemical transport modeling over the TP, particularly in the vertical direction. We expect this dataset can be used to understand the sources and dynamic evolution of air pollutants over the TP, to assess the impacts of chemical forcers on the climate system at multiple scales, and to facilitate the development and improvement of models in cold regions. The potential applications of this dataset include the following. (1) Performances of chemical transport models are commonly poor when applied over the Tibetan Plateau due to complex topography, complex meteorology, etc. This dataset can be used to reduce the uncertainties in these models, especially in the vertical direction (Liu et al., 2021a). (2) This dataset can assist in the source apportionments of atmospheric composition at different altitudes over the Tibetan Plateau. (3) Observed atmospheric composition over the TP is a valuable input for box models to understand atmospheric oxidation capacity at different altitudes on the Tibetan Plateau.