APCC Data Report I: Black carbon and organic carbon dataset from atmosphere, glaciers, snow cover, precipitation, and lake sediment cores over the Third Pole

. The Tibetan Plateau (TP) and its surroundings, also known as the Third Pole, play an important role in the regional and global climate and hydrological cycle. Carbonaceous aerosols (CAs), including black carbon (BC) and organic carbon 25 (OC), can directly/ or indirectly absorb and scatter solar radiation, and change the energy balance on Earth. CAs, along with other atmospheric pollutants (e.g., mercury), can frequently be transported over long distances into the inland Tibetan Plateau. a coordinated monitoring network and research program on Atmospheric Pollution and Cryospheric Change (APCC) has been gradually setup and continuously operated within the Third Pole regions to investigate the linkage between atmospheric pollutants and cryospheric change. This paper presents a systematic dataset of BC, OC, water soluble 30 organic carbon (WSOC), and water insoluble organic carbon (WIOC) from aerosols (20 stations), glaciers (17 glaciers, including samples from surface snow/ice, snowpit, and two ice cores), snow cover (2 stations continuous observed, and 138 locations sites surveyed), precipitation (6 stations), and lake sediment cores (7 lakes) collected across the TP and its surroundingsThird Pole, as the first dataset released from this based on APCC program. These data were created based on online (in-situ) and laboratory measurements. High-resolution (daily scale) atmospheric equivalent BC concentrations were 35 obtained by using an aethalometer (AE-33) in the Mt. Everest (Qomolangma) region, which can provide a new insight into the 2 mechanism of BC transportation over the Himalayas. Spatial distributions of BC, OC, WSOC and WIOC from aerosols, glaciers, snow cover, and precipitation indicated different features among the different regions of the TPThird Pole, which were mostly influenced by emission sources, transport pathways, and deposition processes. Several hundred yearsHistorical records of refractory BC records from ice cores and BC from lake sediment cores revealed the strength of human activities 40 impacts since the industrial revolution. BC isotopes from glaciers and aerosols identified the relative contributions of biomass and fossil fuel combustion to BC deposition on the Himalayas and TPon the Third Pole. Mass absorption cross section of BC and WSOC from aerosol, glaciers, snow cover, and precipitation samples were also provided. This updated dataset is released to the scientific communities focusing on atmospheric science, cryospheric science, hydrology, climatology and environmental science. The related datasets are presented in the form of excel files. These files are available to download from the State Key 45 Laboratory of Cryosphere Sciences, Northwest Institute of Eco-Environment and Resources, Chinese Academy of Sciences at Lanzhou (https://doi.org/10.12072/ncdc.NIEER.db0114.2021, Kang and Zhang, 2021). In the future, datasets of mercury, heavy metals, and POPs will be reported gradually.

Due to the harsh environment and poor accessibility, observed data on the TP are still scares, which limits the effective 75 investigation, verification, and calibration of reanalysis data and modelling simulations (Qian et al., 2015;You et al., 2020).
In particular, studies on atmospheric carbonaceous components (e.g., black carbon, BC; organic carbon, OC) have gained wide attentions, due to the fact that they can directly or indirectly absorb or scatter solar radiation at the surface, alternately interact with the nucleation of clouds, and influence the precipitation efficiency (Bond et al., 2013;IPCC, 2013;Ji et al., 2016;Ramanathan et al., 2005;Ramanathan and Carmichael, 2008;Ramachandran et al., 2020;Yang et al., 2020). These 80 carbonaceous aerosols (CAs) have substantially influenced the climate and environmental changes on Earth Li et al., 2017a;Xu et al., 2012;Chen et al., 2019b;Zhang et al., 2017a;Zhang et al., 2012). The TP has recently been polluted by anthropogenic emissions mainly from long-range atmospheric transport, especially during the pre-monsoon when the aerosols accumulated and combined with westerlies and local mountain valley breeze Painicker et al., 2021;Yang et al., 2018). Although the atmospheric environment over the TP is minimally disturbed by local 85 anthropogenic activities because of the sparse population and limited industries (Cong et al., 2013;, local emissions from booming tourism traffic or domestic fossil fuel/biomass burning also potentially contributed to the climatic and environmental changes . The investigations of CAs from various environmental media in the TP will improve our understanding of recent environmental changes and their impact on rapid cryospheric melting under climate warming. 90 The glaciers provide natural archives of climate and environmental information (Jouzel, 2013;Thompson, 2000;Yao et al., 2006). Historical BC records from the Tibetan glacier ice cores revealed the impact of anthropogenic emissions, with BC concentrations increased by 2−3 times since the 1950s (Kaspari et al., 2011;Wang et al., 2015. Light absorbing BC and OC (including water soluble organic carbon (WSOC) and water insoluble organic carbon (WIOC)) deposited on the glacier and snow cover induced surface darkening and enhanced melting (Flanner et al., 2007;Kang et al., 2020;Lau and Kim, 2018;95 Santra et al., 2019;Xu et al., 2009;Zhang et al., 2020b). Estimates indicated that BC in snow resulted in accelerated glacier 4 melt by approximately 15−20% in the southeast/central TP and Central Asia Zhang et al., , 2020b, and reduced snow cover duration by 3−4 days across the TP and 3−6 days in Northern Xinjiang (Zhang et al., 2018a;Zhong et al., 2019). Meanwhile, WSOC in snow may also contribute to glacier melting (Hu et al., 2018;Gao et al., 2020), due to their light absorption property equal to approximately 10% of that by BC in the northern TP and Northern Xinjiang (Yan et al., 2016;100 Zhang et al., 2019). The isotopic signatures of BC in glacier and modelling results constrained BC sources in the TP, which indicated that BC was predominately derived from fossil fuel combustion (>60%) in the northern TP. Meanwhile in the southern TP, comparable contributions from fossil fuel (46%) and biomass (54%) combustion to BC were observed (Li et al., 2016a;Yang et al., 2018;Zhang et al., 2017a). WSOC in precipitation can provide carbon to the ecosystems .
It was reported that the average WSOC concentrations of precipitation at Nam Co region (inland TP) were lower than that in 105 the urban areas but higher than that from snowpit samples .
As BC is chemically inert in the lake sediment cores, it can serve as an archive and reliable indicator to investigate the source and transport of BC in the past (Han et al., 2012). The BC investigation from lake sediment cores over the plateau revealed an increasing trend of BC concurrent with increased anthropogenic emissions since the 1950s (Cong et al., 2013;Han et al., 2015;110 Neupane et al., 2019;Zhu et al., 2020), and suggested that BC deposition in recent decades have increased about 2−3 fold compared to the background level.
During the past decade, our research team has gradually setup a coordinated monitoring network and research program to link atmospheric Atmospheric pollution Pollution and c Cryospheric change Change (APCC) covering the TP and its surrounded region ( Fig. 1)  . Based on the APCC program, our overarching goal is to perform more integrated 115 and in-depth investigations of the origins and distributions of atmospheric pollutants and their impacts on cryospheric change.
The updated specific goals included: (I) Characterize the features of atmospheric pollutants (including new emergent pollutants) and depict their spatial and seasonal variations in different environmental media over the Third Pole region; (II) Investigate the source regions appointment of different atmospheric pollutants based on chemical tracers and modelling, 120 and reveal the transport pathways and mechanisms by which atmospheric pollution is trans-boundary transported to the Third Pole region; (III) Quantify the contribution of atmospheric pollutants deposited as light-absorbing impurities to the glacier and snow melting, determine the fates of environmentally relevant pollutants within glaciers and snowpack, and further estimate the feedback of cryospehric melting to the carbon (and nitrogen) cycle and hydrological changes. 125 During the past 10 years, a A series achievement has been published on the research progress that depends on or is related to the APCC network during the past decade . Recently, new progresses were also achieved, including extended the study area to central Asia Zhang et al., 2021a), investigated a new emergent pollutantmicroplastics in snow , and discussed the potential impact of glacier melting and permafrost thaw on carbon cycle (Gao et al., 2021a;Zhang et al., 2021c). Therefore, in this article, we will introduce and provide access to the 130 5 systematic dataset of BC and OC from the atmosphere, glaciers (including ice cores), snow cover, precipitation, and lake sediment cores over the TP and its surroundings based on the APCC network. The site description, online observations, and sampling are summarized in Section 2 and 3, respectively. A detailed data description of laboratory analysis and BC, OC, WSOC, and WIOC concentrations from aerosols/glaciers/precipitation/snow cover, and refractory BC (rBC) historical records from ice cores and BC from lake sediment cores are given in Section 4, highlighting the primary results and differences and 135 similarities among studied regions. Meanwhile, mass absorption cross section (MAC) values of BC and WSOC from aerosols/glaciers/snow cover and precipitation, and carbon isotopic signatures (∆ 14 C, δ 13 C) from glacier snowpits/precipitation and aerosols are also provided. The data availability and access are provided in Section 5, and the conclusions are summarized in Section 6.

Overview of site distributions
The TP can be divided into three distinct domainssub-regions, respectively associated with the dominant influence of the westerlies (northern TP), Indian monsoon (southern TP), and the transition region in between, which don't have an exact boundary of each domain (Yao et al., 2013). These different domain features in sub-regions motivated the need for network observations to understand the atmospheric pollutants and their possible impact on the environment. The APCC program 145 currently consists of 29 stations (and sites) across the TP and its surroundings covering these three distinct domainssub-regions.
Among these stations, there are 27 stations for aerosol sampling and observations, 20 glaciers observed, 3 stations for snow cover observations and 138 locations for snow sampling once (138 sites surveyed), and 6 stations for precipitation observations (Table 1 and Fig. 1). In addition, sediment cores were studied from 7 lakes across the Himalayas and TP. According to the distance and the extent of impact from anthropogenic activities, these sites stations were distributed into two major types, 150 namely urban sites stations (strongly influenced by anthropogenic activities), and remote sites stations (weakly impacted by direct anthropogenic emissions). As this study paper is the first dataset report based on the APCC program, we will release the carbonaceous dataset from 20 stations for aerosols, 17 glaciers (including samples from surface snow/ice, snowpit and two ice cores), 2 stations for continuous snow cover observations and 138 sites 138 locations with during snow cover surveys, 6 stations for precipitation, and 7 lake sediment cores across the TP and its surroundings. In the future, more comprehensive 155 datasets on mercury, heavy metals, and persistent organic pollutants will also be reported gradually from the APCC program.

Stations for the aerosol and precipitation studies
The APCC network has a total of 29 stations for the aerosol and precipitation studies. In specific, Ffor the spatial distributions ( Fig. 1 (Dushanbe, Toskent, Bishkek, Jimunai, Koktokay, Tianshan, Koxkar, Muztagh Ata, Ngari, Laohugou, Qilian, Beiluhe) distributed in Central Asia, Xinjiang Uygur Autonomous Region (China) and the western and northern TP, mainly controlled by the westerlies. In the inland TP, 2 stations (Nam Co, Tanggula) are distributed across the Nainqentanglha Mountains. to the Tanggula Mountains, which are alternately influenced by the Indian monsoon and westerlies (as the transition domain, Yao et 165 al., 2013). Aerosol samples were also collected from Lhasa and Lanzhou city. The elevations of these stations ranged from 13 m to over 5000 m a.s.l. (Table 1), with landscapes including forest, alpine steppe, alpine meadow, and desert. These sites served as the key locations for field observations and measurements.
There are 5 stations located on the southern side of the Himalayas (located in Nepal), focusing on aerosol studies to resolve the transboundary transport of air pollutants (Tripathee et al., 2017;Chen et al., 2020). A list of the detailed information 170 on observation items is in Table 1. Kathmandu, the capital city of Nepal, is characterized by rapid but uncontrolled urban growth and has severe air pollution problems. Pokhara, a famous tourist city, has undergone rapid urbanization with increased numbers of vehicles and industries. Lumbini, located on the northern edge of the Indo-Gangetic Plain (IGP), is a typical rural site located in a mixed setting with a large number of agricultural and industrial activities. Dhunche is a small town situated in the Langtang National Park in the Rusuwa district in the foothills of the Himalayan Mountains; it is approximately 50 km north 175 of Kathmandu. Jomsom is a semi-arid small town in the Mustang district, located in the Kali Gandaki River Valley across the Himalayas.
Karachi is the capital city of Sindh province, with almost half of the domestic industries in this city. The sampling place is in the extreme south portion, just a few kilometres away from the Arabian sea. Mardan, located near Peshawar city, is the second In the Central Asian countries, 3 stations were selected and set up for the aerosol studies. Dushanbe is the capital city of 185 Tajikistan. The aerosol sample collection was performed at the Physical Technical Institute of the Academy of Sciences of Tajikistan, which is located in an urban environment on a hill in the eastern part of Dushanbe. Toshkent Shahri is the capital city of Uzbekistan, located in the east part of the country and between the Tianshan Mountains and the Syr river. Bishkek is the capital of Kyrgyzstan, at the foot of the Alatao mountains in north of the country and in the central Chu River basin. All of these sites are located in the arid and semi-arid regions with a dry continental climate. 190 There are 18 stations (Lanzhou, Lhasa, Everest, Zhongba, Nyalam, Lulang, Yulong, Nam Co, Tanggula, Beiluhe, Qilian, Laohugou, Ngari, Muztagh Ata, Koxkar, Tianshan, Koktokay and Jimunai) continuously observed over the Chinese TP and its surroundings within China. Lanzhou is the capital city of Gansu province and is an important industrial base and comprehensive transportation hub in northwest China. Lhasa, China's highest altitude city, is located on the banks of the Lhasa River and serves as the capital of the Tibet Autonomous Region. Everest, Zhongba, and Nyalam are located on the northern 7 side of the Himalayas, characterized by agriculture and yak husbandry and dominated by the Indian monsoon. In particular, at the Qomolangma Atmospheric and Environmental Observation and Research Station of CAS on the north slope of Mt. Everest (Everest station), real-time data of equivalent BC concentrations (eBC, measured by the Aethalometer model, AE-33) has been observed since May of 2015 . Lulang is located in a subvalley of the Yarlung Tsangpo Grand Canyon, a corridor for the warm-humid Indian monsoon to penetrate the inner TP. Yulong, the southernmost glaciated mountain in the 200 Eurasian continent, close to Lijiang City (Yunnan province), may be affected by local emissions. Nam Co is a typical pastoral area in the inland TP. Beiluhe is located to the east of Kekexili of inland TP. Laohugou is a remote hinterland site located in the western Qilian Mountains. Ngari is located in the west TP, a typical arid area covered by bare soil or grasslands. Koxkar station is located on the southern slope of Mt. Tomur, which is the highest peak in western Tianshan. Tianshan Glaciological Station (Tianshan station) is located in the upper reaches of the Urumqi river. Jimunai is located in Northern Xinjiang, near 205 the Altai Mountains.
Precipitation samples in this study paper were collected from 6 stations ( Fig. 1), namely the Upper Heihe river basin (Qilian), Nam Co, Everest, Lulang, Yulong, and Lhasa city Li et al., 2018b;Niu et al., 2019). The Upper Heihe river basin is a typical permafrost basin, located in the Qilian Mountains of the northern TP . The other 5 stations have been described in the previous paragraphs. 210

Glaciers
Carbonaceous data from a total of 17 glaciers are provided in this study paper (Table 2, Fig. 1). There are 11 glaciers located in the Indian monsoon dominated region. Among them, Baishui glacier No.1 of Yulong Snow Mountain and the other 4 glaciers (namely Demula, Renlongba, Yarlong, Dongga) are located in the southeast TP (Niu et al., 2018a;Zhang et al., 2017a), and East Rongbuk glacier is located in the central Himalayas on the northern slope of Mt. Everest. In the inland TP, 3 glaciers are 215 observed and studied. The Zhadang glacier, with an area of 2.0 km 2 and a length of 2.2 km, is located on the north-eastern slope of the Nyainqentanglha mountain range (Li et al., 2018c). The Xiaodongkemadi glacier, with an area of approximately 1.60 km 2 and facing southwest, is located at the headwaters of the Dongkemadi River, a tributary at the upper reaches of the Buqu River near the Tanggula Pass in the central TP Li et al., 2017b). The Ganglongjiama glacier (also known as Guoqu glacier) is located on the northern slope of Mt. Geladaindong, the summit peak of the Tanggula Mountains 220 (Hu et al., 2020a). Meanwhile, there are 5 glaciers studied in northern Pakistan. The Passu and Gulkin glaciers are located very near the Karakoram highway connecting Pakistan with China. The Barpu and Mear glaciers are located very close to each other and around 3 km away from the residential area of the Hopar and Nagar valleys and Sachin glacier is close to a small city (Astore) (Gul et al., 2018).
There are 6 glaciers monitored in the westerly-dominated regions. The Laohugou glacier No. 12, a typical valley glacier, 225 is located on the north slope of the western Qilian Mountains in the northern TP and covers an area of 21.9 km 2 (Li et al., 2019a;Zhang et al., 2017b). The Muztag Ata glacier has an area of ∼0.96 km 2 and a length of 1.8 km, located in the eastern Pamir Plateau (Yao et al., 2012). The Anglong glacier is located in the headwater region of Indus, covering an area of 1.5 km 2 8 (Chen et al., 2019c). The Koxkar glacier is located on the south slope of Mt. Tomur, the highest peak in western Tianshan, on the border between China and Kyrgyzstan (Zhang et al., 2017c). The Urumqi glacier No.1 is located at the headwater of the 230 Urumqi River in eastern Tianshan, which is surrounded by the Taklimakan Desert to the south, the Gurbantungut Desert to the north, and the Gobi Desert to the east. This glacier has an area of 1.65 km 2 (50% of the basin) (Li et al., 2019b). The Muz Taw glacier is located on the northern slope of the Sawir mountains in Northern Xinjiang, south of the Ertix river in Central Asia (Zhang et al., 2020b).
There are 2 ice cores retrieved from the East Rongbuk glacier (Mt. Everest region) and the Ganglongjiama (Mt. 235 Geladaindong region) glacier, respectively (Jenkins et al., 2016;Kaspari et al., 2011;Zhang et al., 2015) (Fig. 1). In this study, data from the Muztag Ata glacier and the Urumqi glacier No.1 were not provided, which can be referred to the previous studies Wang et al., 2015).

Snow cover sites
Snow cover samples were collected from 2 stations (the Laohugou and Koktokay) ( Fig. 1) and 138 snow survey sites locations 240 once in the TP and the Northern Xinjiang (Zhang et al., 2018a;Zhong et al., 2019Zhong et al., , 2021. The Koktokay Snow Station at the headwater of the Irtysh River was selected as the fixed-point site, which is located in the Kayiertesi river basin, the first tributary of the Irtysh river. The Kayiertesi river basin is in the southern Altai Mountains in China, covering 2365 km 2 . Seasonally frozen soil and permafrost are widely spread in the basin, and the basin is rich in vegetation (39.2% vegetation coverage in 2014) (Zhang et al., 2016). The land surface is generally covered by forest in the shade and semi-shade and by 245 grassland and shrubs on the sunny and semi-sunny slopes (Zhang et al., 2016). The lowest air temperature is below −45 ℃ in winter. Average annual maximum snow depth exceeds 1 m (Zhang et al., 2014). Floods in the river basin are due to rain-onsnow events, originating from the combination of rapidly melting snow and intense precipitation. There were 27 sites locations for snow cover sampling distributed in the southern TP, 10 sites in the central TP, and 10 sites from one glacial river basin (Laohugou region in the northern TP) (Zhang et al., 2018a). There were 91 locations sites distributed across Northern Xinjiang. 250 Among them, 11 sampling locations sites were selected during the whole snow season, including 5 locationssites located in the southern Altai Mountains, 3 locationssites on the west side of the Junggar Basin, and 3 locationssites in the northern Tianshan Mountains (Zhong et al., 2019).

Sediment cores from the lakes
Lake sediment cores were studied from 7 lakes distributed across the Himalayas and TP (Cong et al., 2013;Neupane et al., 255 2019) ( Table 3, Fig. 1). Gosainkunda and Gokyo are located on the southern slope of Nepal Himalayas. Qiangyong and Ranwu lake are located in the southern and southeast TP, respectively. Nam Co, Lingge Co and Tanggula lake are located in the inland TP. Riverine BC inputs to Qiangyong and Ranwu lake were controlled by the surrounding snow cover and glacier meltwater.
All the above lakes are located far from areas with large amounts of anthropogenic activities. 9 3 Sampling in the field 260

Atmospheric aerosol and precipitation sampling
In this study, carbonaceous data (EC, OC, WSOC) from atmospheric aerosols (TSP, total suspended particulate) were collected from 19 sites stations based on the APCC program, distributed in the TP, Southeast Asia, and Central Asia. Generally, the TSP were collected on pre-combusted (550 º C, 6 hours) quartz fibre filters (90 mm in diameter, with pore size of 2.2 μm, Whatman) with a TSP cyclone at a flow rate of 100 L/min for 24 hours (urban or rural sites) or 48 hours (remote sites) (Chen et al., 2019b). 265 The TSP sampler was setup on the roof of the observation building to avoid the effects of ground dust (Fig. 2). After sampling, the filters were kept frozen until analysis. TSP samples were collected every 6 days to bypass the "weekend effect", which indicated that weekly cycle of aerosol composite (with low value sin weekend and high values in the weekdays) was usually governed by the anthropogenic emissions (Satheesh et al., 2011). These quartz filters will be used for measurement of EC, OC, and WSOC from the aerosols. 270 When precipitation occurred, samples were collected by an automated precipitation collector for the analysis of BC and WIOC. After the precipitation event, the precipitation amount was also recorded, and the samples were transferred into HDPE bottles (250 mL) and kept frozen until analysis. Samples of WSOC were collected in prebaked aluminium basins (450 °C, 6 h) that were placed on a 1.5 m high platform (Li et al., 2016b(Li et al., , 2016cNiu et al., 2019). Due to limitations in the volume of samples collected during small precipitation events, only those with enough sample amount were selected to 275 determine ∆ 14 C and δ 13 C.

Glaciers, snow cover and ice core sampling
Surface snow/ice and snowpits were sampled from glaciers, and meteorological variables in glacial regions were observed by Automatic Weather Stations (Fig. 2). Generally, for glacier surface snow/ice sampling, different types of snow/ice (fresh snow, aged snow and granular ice) samples were collected across the ablation and accumulation zone of the entire glacier (Fig. 3). 280 Whirl-pak bags were used to collect surface snow samples from the upper 0-10 cm (or 0−5 cm) of depth (approximately 2 L, unmelted), as well as surface granular ice samples. In general, the snowpit samples were collected from the accumulation zone of glaciers using a stainless-steel spoon with a vertical depth interval of 5, 10 or 15 cm and transferred into a Whirl-pak bag, following the protocol described by Kang et al. (2007). Generally, duplicate samples were collected to estimate the differences between sampling. In situ observations included snow thickness, density, grain size, and surface albedo, which are used to 285 estimate the post-deposition processes, chemicals flux, and effect of BC on glacier melt.
In field surveys across the northern/eastern/southern TP and Northern Xinjiang (Zhang et al., 2018a;Zhong et al., 2019), snow depth, snow density, snow grain size and surface albedo were observed, and snow cover samples were collected. At each site, snow samples were collected from the top 5 cm of the snowpack and stored in a Whirl-pak bag. The vertical resolution of the snowpit profile at Koktokay snow station was 5 cm intervals from the snow surface to the depth of 20 cm, then samples 290 were collected every 10 cm from the depth of 20 cm to the bottom. During the snow accumulation period (November) and the stable period (from December through early March), snowpit samples were collected three times a month at 10:00 am (UTC/GMT +8.00, the same below). During the intense snowmelt period (from early or mid-March to the early April), snow samples were collected twice a day at 10:00 am and 7:00 pm, respectively. Snow cover samples for analyses of WSOC were directly collected into the square polycarbonate bottles . The samples were kept frozen until they were 295 melted and filtered in the laboratory.
In this studypaper, historical rBC records data from two ice cores were reported. In 2002, a 108 m ice core was collected from the col of the East Rongbuk glacier located on the northeast ridge of Mt. Everest on the northern slope of the Himalayas (Kaspari et al., 2011). In 2016, a shallow ice core (8 m length) was drilled by the team from the same glacier to expand rBC records since 2000. In November 2005, a 147 m ice core was collected from the upper basin of the Ganglongjiama glacier on 300 the northern slope of the Mt. Geladaindong using an electro-mechanical drill (Jenkins et al., 2016;Zhang et al., 2015). The drilled ice cores were packed in polyethylene tubing in the field, transported frozen to the State Key Laboratory of Cryospheric Science, Chinese Academy of Sciences in Lanzhou, and kept in a cold room at -20°C until sample preparation and analysis .

Lake sediment core sampling 305
Lake sediment cores were drilled from the deep basin of the studied lakes during 2008−2017 using a gravity coring system with a 6 cm inner diameter polycarbonate tube (Cong et al, 2013;Neupane et al., 2019). The core sediments were sliced in the field at intervals of 0.5 cm, except for Lingge Co and Ranwu Lake, which were sliced at 1 cm intervals, stored in plastic bags, and kept frozen until analysis.

Real-time atmospheric BC using Aethalometer
BC is an important part of atmospheric particulate aerosols, which imposes adverse effects on atmospheric visibility, health, and climate change (Ramanathan and Carmichael, 2008). The eBC is operationally defined as the amount of strongly lightabsorbing carbon with the approximate optical properties of carbon soot that would give the same signal in an optical instrument (e.g., Aethalometer) (Andreae and Gelencser, 2006). At the Everest station, a real-time optical instrument 315 (Aethalometer model AE-33, Magee Scientific Corporation, USA) for measurement of eBC from atmospheric aerosols was operated with an inlet installed at approximately 3 m above the ground level since March of 2015. The airflow rate was operated at 4 L min −1 . The eBC concentrations can be acquired according to the light absorption and attenuation characteristics from the seven fixed wavelengths (e.g., 370, 470, 520, 590, 660, 880 and 950 nm) Drinovec et al., 2015). In general, the eBC concentration measured at 880 nm is used as the BC concentration in the atmosphere, as the absorption of 320 other types of aerosols (e.g., OC and dust) is greatly reduced at this wavelength (Sandradewi et al., 2008). When calculating eBC concentrations, it is therefore possible to eliminate the "loading effect" with the loading compensation parameter k, which 11 allows extrapolation to zero loading, and the accurate ambient BC concentration is obtained (Drinovec et al., 2015). Previous studies have evaluated the real-time compensation algorithm of dual-spot Aethalometer model AE-33 and indicated that AE-33 agreed well with the post-processed loading effect compensated data obtained using earlier Aethalometer models and other 325 filter based absorption photometer Drinovec et al., 2015).
At Everest station, daily and monthly mean eBC concentrations presented a strong seasonal pattern variations during 2015−2019, which showed the highest values in the pre-monsoon season (~923 ng m −3 in April) and the lowest values in the monsoon season (~88.5 ng m −3 in July) (APCC dataset I-1 and Fig. 4). In the pre-monsoon season, BC from the Indo-Gangetic Plain (IGP) region can be transported and concentrated on the southern slope of the Himalayas by the north-westerly winds in 330 the lower atmosphere, and then further transported across the Himalayas by mountain-valley winds .
Simulations further indicated that the BC aerosols in South Asia could be uplifted and transported to the Mt. Everest region by the southerly winds in the upper atmosphere in the monsoon season . The results indicated the seasonal cycle of BC was significantly influenced by the atmospheric circulation and combustion intensity in the Mt. Everest region.

Analysis methods and data of atmospheric aerosol EC and OC
After the TSP sampling, the aerosol EC and OC concentrations were measured by a thermal/optical carbon analyser (Desert Research Institute DRI Model 2001 or Sunset Lab) (Chen et al., 2019b). In detail, for the aerosol samples (TSP quartz filters), 340 a punch of sample (area 0.5 cm 2 ) was put in a quartz boat inside the analyser and heated stepwise at the different temperature plateaus (IMPROVE_A temperature protocol with an optical reflectance correction for sample charring) .
The IMPROVE_A temperature protocol defined temperature plateaus for thermally derived carbon fractions of 140 °C for OC1, 280 °C for OC2, 480 °C for OC3, and 580 °C for OC4 in a helium (He) carrier gas and 580 °C for EC1, 740 °C for EC2, and 840 °C for EC3 in a 98% He/2% oxygen (O2) carrier gas. Each carbon fraction reported to the IMPROVE_A network 345 database consists of a value and a precision. The aerosol OC and EC by thermal/optical reflectance (TOR) are insensitive to the change in such temperature protocol. Therefore, EC, OC, and total carbon (TC) were calculated from the eight carbon fractions as: Where, OP in the equation represents pyrolyzed OC, which is defined as the carbon evolving between the introduction of oxygen in the helium atmosphere and the return of reflectance to its initial values (the OC/EC split) (Chow et al., 2005). The accuracy of the measurement was ±10% and the detection limit for OC, EC and TC were 0.43, 0.12, and 0.49 μg C cm −2 , respectively (Chen et al., 2019b).

12
Approximately one thousand1,000 samples from 19 different sites were analysed in this study (APCC dataset I-1). Annual average TC, EC and OC distributions across the TP usually showed higher values in the urban and rural sites and lower values in the remote sites, which considerably decreased from outside to inland of the TP (Chen et al., 2019b) ( Figure Fig. 5 primarily during the pre-monsoon season, due to regional-scale pollution plumes known as atmospheric brown clouds (Ramanathan et al., 2007b). However, different seasonal variations were observed in the inland to northern TP. For example, at Nam Co station, relatively high OC and EC concentrations occurred during the monsoon and post-monsoon. The regional differences in CA seasonal variations suggests differences in the pollutant sources and transport pathways. In the Central Asia,

the OC and EC concentrations demonstrated clear seasonal patterns, with elevated concentrations observed during August to 370
February . These results of different seasonal variations between southern and northern parts of the TP, suggesting differences in the patterns of pollutant sources and in distance from the sources between the regions (Chen et al., 2019b). All of these measurements have provided the basis for understanding the spatio-temporal variations of carbonaceous particles over the vast Third Pole region, which is of great importance for scientific communities worldwide. Furthermore, these data sets are critical for further scientific studies in the future on the atmospheric environment across this vast region. 375

Mass absorption cross section (MAC) of aerosol EC and WSOC from atmospheric aerosols
The optical attenuation (ATN) can be calculated based on the transmittance signal with equation (4) (Cheng et al., 2011): Where, Tinitial and Tfinal are the transmittance signal before and after the thermal-optical analysis, respectively. The ATN can be used to determine the absorption coefficient (babs) based on equation (5): 380 babs= ATN × Where, A is the filter area (cm 2 ) and V is the volume of air sampled (m 3 ). Thus the MAC of EC (MACEC, m 2 g −1 ) can be calculated as: Where, ECS (μg C m −2 ) is the filter loading of EC. As multiple scattering effects occur associated with the filter-based 385 measurement of absorption, they were corrected by a value of 3.6 (Chen et al., 2021).

13
The light absorption spectra of WSOC were measured between wavelengths of 200 nm to 800 nm with precision of 5 nm bandwidth by a UV-Visible Spectrophotometer (SpectraMax M5). The value of MACWSOC can be calculated from following equation (7) (Bosch et al., 2014;Kirillova et al., 2014): × ln (10) (7) 390 Where, A (absorbance) is derived directly from the spectrophotometer, L is the absorbing path length, and c is WSOC concentration.
The MACEC at 632 nm exhibited significant spatial variations, differing by a factor of up to two with a clear increasing trend from the outer to the inner TP (Fig. 6). For the study sites, the annual MACEC ranged from 6.37 to 8.49 m 2 g −1 . Karachi had the lowest MACEC among the sites. Other sites, including Mardan, and Lanzhou, exhibited similar annual average MACBC 395 of approximately 7.0 m 2 g −1 during the sampling period. On the southern side of the TP, sites (e.g., Dhunche, Jomsom, Everest, Zhongba, Nyalam, Lulang) had moderate MACEC, which were slightly higher than those in urban sites but lower than in the inland remote regions (e.g., Nam Co, Hunza, Ngari, and Beiluhe). respectively. The MACWSOC values were generally higher in urban regions than those at remote sites probably because of the larger primary and anthropogenic contribution for WSOC in urban areas in contrast to remote regions .

BC and WIOC from glaciers and snow cover
Before filtration, these frozen snowpit, surface snow, and snow cover samples were rapidly melted via a hot water bath (approximately 20 minutes for complete melting) and the meltwater (typically 0.5 to 1 L) was filtered through a pre-dried (in 410 a desiccator, at 550°C, for 6 hours) weighted quartz filter (Whatman, with pore size of 2.2 μm) using a vacuum pump (Zhang et al., 2017aZhong et al., 2019Zhong et al., , 2021. Samples were filtered twice and the filtration equipment was rinsed with ultrapure water twice (<18.2 mΩ) in order to prevent particle loss. The estimated total uncertainty in the particle concentrations was <1% (including background counts and random counting errors).
After filtering the meltwater, the quartz filters were dried and weighed gravimetrically, then analysed for BC and WIOC. 415 Generally, the dust loads in the snow/ice samples were greater (approximately 2−3 orders of magnitude) than in the airborne aerosol samples. In order to eliminate the impact of dust, BC and WIOC were measured by a modified IMPROVE_A protocol by DRI model thermal/optical carbon analyser (Yang et al., 2015;Wang et al., 2012). Specifically, the method was modified such that only one temperature plateau (550°C) was used in the 100% helium atmosphere to reduce the time that the BC was exposed to the catalysing atmosphere. The reported OC concentrations from the snow and ice samples account for only WIOC 420 because most of WSOC was not captured by the filter-based method. The detection limit of the analysis was 0.19 ± 0.13 μg TC cm −2 , and the filter blank was 1.23 ± 0.38 μg TC cm −2 , which was about 1 order of magnitude lower than the measured sample values .
The evaluated blank filters for total carbon <1 μg cm −2 . For the same filter, multiple measurements showed small relative standard deviations (RSD, <10%), indicating that the data points tended to be close to the mean value, an 425 acceptable filtration. The duplicate snow samples demonstrated the similar concentrations of BC and OC. The evaluated the impact of inorganic carbonates interfering with BC measurements showed the carbonate acidification and analysis indicated acceptable data quality with a discrepancy <20% (Zhang et al., 2017a).
According to the measurements, the average BC concentrations in glacier snow/ice ranged from several ng g -1 to hundreds of ng g -1 , with marked differences between the glaciers (APCC dataset I-2, and Fig. 7). In particular, BC concentrations in the 430 aged snow/granular ice were usually much higher (1−2 orders of magnitude) than that in fresh snow/snowpit/ice cores . Concentrations of BC and WIOC were higher in the central and northern TP than in the southern TP. There was large spatial variability of BC and WIOC in snow cover across the TP and Northern Xinjiang. Concentrations of BC and WIOC in snow cover over the TP were 202−17468 ng g −1 and 491−13880 ng g −1 , respectively. The values of BC and WIOC in snow cover across Northern Xinjiang varied from 32 to 8841 ng g −1 , and 77 to 8568 ng g −1 , respectively. Greater BC and WIOC 435 concentrations in snow cover appeared in the western areas (west of 83°E) than other areas in Northern Xinjiang (Fig. 8).
Vertical variations of monthly mean BC and WIOC concentrations in the snowpit profiles showed that the maximum monthly mean BC and WIOC concentrations generally appeared at the snow surface (302−6271 ng g −1 for BC and 780−17877 ng g −1 for WIOC) (Fig. 9), suggesting that the magnitude of downward migration of BC and WIOC was much less than the enrichment in surface snow for snow cover. 440

WSOC from glaciers and snow cover
Before the measurement, the melted snow/ice samples (collected using pre-cleaned 60 mL square-shaped polycarbonate bottles) were filtered through a 0.45 μm (pore size) PTFE membrane filter (Macherey-Nagel). WSOC concentrations were determined using a TOC-5000A analyser (Shimadzu Corp, Kyoto, Japan) (Yan et al., 2016;Zhang et al., 2018b). The detection limit of the instrument, precision and average WSOC concentration of the blanks were 0.015 mg L -1 , ±5% and 0.025 ± 0.006 mg L -1 , 445 respectively, demonstrating that contamination was minimal during the pre-treatment and analysis processing. WSOC concentrations in glacier samples were lower in the snowpit and fresh snow, but higher in the aged snow or granular ice (Fig.   7). For the benchmark glaciers (Muz Taw glacier, and Laohugou Glacier No.12), the spatial distribution of WSOC generally decreased with increasing elevation, indicating that more intense melting occurred in the lower elevation ablation zones exposed to higher concentrations of WSOC Hu et al., 2018). The detection limit of the analyser was low at 450 15 4 μg L −1 , while the precision and average WSOC concentrations of the blanks were ± 5% and 4 ± 2 μg L −1 , respectively, demonstrating that contamination during the pre-treatment and analysis processing of these samples was weak .
The light absorption spectra of the WSOC were measured with a UV-Visible Spectrophotometer (SpectraMax M5, Molecular Devices, USA) between the wavelengths of 200 nm and 800 nm with an interval of 5 nm. Each spectrum was determined relative to that of Milli-Q water. The MAC values for WSOC in glaciers and snow cover samples were calculated 455 by equation (7). The MACWSOC from snowpits of Laohugou glacier No.12 and Ganglongjiama glacier (Central TP) was 4.71±3.68 m 2 g −1 and 2.17±2.13 m 2 g −1 , respectively (Hu et al., 2018(Hu et al., , 2020. MACWSOC from snow cover in the Altai mountains was 0.45 ± 0.35 m 2 g −1 , with higher values in March and April 2017; the fraction of radiative forcing caused by WSOC relative to BC accounted for approximately 10.5%, indicating WSOC was a non-negligible light-absorber in snow of the Altai regions . The calculated MACWSOC from Baishui glacier No.1 at Mt. Yulong was 6.31±0.34 m 2 g −1 (Niu et al., 460 2018b). The comparisons of MACWSOC from glaciers and snow cover revealed significant differences between regions (Fig.   10). Even for the same glacier (the Baishui glacier No.1), MACWSOC showed large variability from snowpits over different sampling periods (Niu et al., 2020).

WSOC from precipitations
The precipitation samples were also filtered through a PTFE membrane filter with 0.45 μm pore size (Macherey-Nagel) before 465 the WSOC measurement. The Shimadzu TOC-5000 total organic carbon analyser (Shimadzu Corp, Kyoto, Japan) was used to determine the precipitation WSOC concentration (Li et al., 2016d). The average blanks for precipitation WSOC was 0.08± 0.05 μg C mL −1 . The light absorption spectra of precipitation WSOC were measured by using the same method used for the glacier samples. The value of precipitation MACWSOC was calculated by the equation (7).
Precipitation WSOC concentrations decreased from urban cities to remote stations, with marked seasonal variations 470 (APCC dataset I-3, and Fig. 11) Li et al., 2017aLi et al., , 2018aNiu et al., 2019). The average precipitation WSOC concentrations was 1.41 μg C mL −1 at the Qilian Station of the northern TP, with WSOC flux of 6.42 kg ha −1 yr −1 . For summer precipitation in the Mt. Yulong region in southeast TP, the average concentration of WSOC was 1.25 μg C mL −1 (Niu et al., 2019). The average precipitation WSOC concentration at Nam Co was 1.0 μg C mL −1 ; the estimation suggested that about15% of WSOC was fossil derived . The MACWSOC values of precipitation samples were 475 significantly lower than the aerosol samples (Li et al., 2017a). For example, the precipitation MACWSOC only ranged 0.26-1.84 m 2 g −1 at Qilian Station, suggesting the potential impact of WSOC on climatic forcing in the area. Seasonally, MACWSOC of both aerosol and precipitation samples showed high values in winter and low values in summer Li et al., 2021).

Carbon isotopes from glacier snowpits and atmospheric aerosols 480
The dual-carbon-isotope signatures (∆ 14 C and δ 13 C) are effective ways to distinguish and track the different sources of carbon aerosols (Gustafsson et al., 2009). In this study, BC mass contents of the filtered glacier snowpit and atmospehric aerosols (TSP) samples were quantified using the TOT technique. Briefly, filters were acidified by fumigation in open glass Petri dishes held in a desiccator with >37% HCl acid for 24 hours to remove carbonates and were subsequently dried at 60 ℃ for 1 hour to remove remaining HCl acid. A 1.5-cm 2 punched acid-treated filter was analysed using a carbon analyser (Sunset Laboratory,485 Tigard, OR) following the National Institute for Occupational Safety and Health (NIOSH) method 5040 to determine BC and OC concentrations of the aerosol samples and the WIOC concentrations of the particles from the collected snow samples.
Sucrose standard and other reference materials were also subjected to these measurements (Li et al., 2016). The filter area required for the subsequent 14 C measurements was determined based on the measured BC concentration. The CO2 produced was cryotrapped during the BC combustion phase after removing the water and sulfur-containing gases online. Purified CO2 490 was then transferred in flame-sealed glass ampules to the United States National Science Foundation (US-NSF) National Ocean Science Accelerator Mass Spectrometry (NOSAMS) facility at the Woods Hole Oceanographic Institution (Woods Hole, MA, USA). For precipitation and snow samples containing more than 60 μg C (WSOC), the sample was poured into pre-cleaned quartz tubes, acidified to pH 2 with phosphoric acid, and sparged with ultrahigh purity helium to remove inorganic carbon. Next, the sample was irradiated using a high-energy UV lamp for 5 hours to quantitatively oxidize WSOC to CO2. 495 Concentrations of WSOC were determined using a calibrated Baratron absolute pressure gauge (MKS Industries). WSOC concentrations determined by this method well agreed with those determined using the Shimadzu TOC analyser (Raymond et al., 2007). CO2 was cryogenically purified by liquid nitrogen on a vacuum extraction line and sent to the National Ocean Sciences Accelerator Mass Spectrometry (NOSAMS) at Woods Hole for isotopic analysis.
Fossil fuel contribution to BC in both aerosol and snowpit samples decreased from the outer TP (i.e., South Asia and 500 western China) to the inner TP (APCC dataset I-4, and Fig. 12) (Li et al., 2016). For the Himalayan region (Thorung, East Rongbuk, Qiangyong), equal contributions from fossil fuel (46±11%) and biomass (54±11%) combustion were found, consistent with BC source fingerprints from the IGP region (Dasari et al., 2020). BC in the remote northern TP (Laohugou) predominantly derived from fossil fuel combustion (66±16%), consistent with Chinese sources. The fossil fuel contributions to BC in the inland TP (Tanggula, Zhadang) were lower (30±10%), implying contributions from internal Tibetan sources (for 505 example, yak dung combustion). A similar phenomenon was observed for WSOC in snowpits, which revealed lower fossil fuel contributions in the inner TP. The results suggested that pollutants transported from South Asia influenced the BC and WSOC concentrations over the southern part of TP. Meanwhile, Chinese emissions influenced northern TP and local emissions influenced inland TP.

rBC from ice cores 510
In the laboratory, the drilled ice core sections were cut longitudinally into halves, which were sectioned using a modified bandsaw (stainless-steel blades; table tops and saw guides covered with Teflon) cleaned with ethyl alcohol and ultrapure water (18.2 MΩ). One half-section of the core was cut into 3−5 cm segments. For the East Rongbuk ice core, a total of 663 samples were collected for rBC analysis (Kaspari et al., 2011). For Geladaindong ice core, a total of 3585 samples were collected for rBC analyses (Jenkins et al., 2016). To eliminate possible contamination from sampling, drilling or storage, the outer portion 515 of each ice sample was scraped using a ceramic knife in a clean, low-temperature room (-8 °C). Polypropylene clean roomsuits and non-powder vinyl clean-room gloves were worn throughout the sampling process to minimize potential contamination (Kang et al., 2007). After the outer section of the ice was removed, the samples were put into Whirl-pak bags, melted at room temperature, and the solution was poured into high-density polyethylene vials for the subsequent experiments.
The samples were stored as liquid until analysis (Jenkins et al., 2016). 520 The rBC records from ice cores were analysed by using a SP2 (Droplet Measurement Technologies), an important analytical technique applied to Arctic ice cores (McConnell et al., 2007). The SP2 uses laser induced incandescence to measure the refractory BC mass in individual particles quantitatively and independent of particle morphology and coatings with light scattering material, with detection limit was 0.3 fg/particle (Schwarz et al., 2012). Coupled to a nebulizer, the SP2 can be used to measure rBC particles in liquid-phase samples (Lim et al., 2014;Schwarz et al., 2013). The high sensitivity and small-525 required sample volume enabled ice cores to be analysed at much higher resolution than thermal-optical methods. Generally, rBC concentrations from ice cores are likely to be systematically underestimated due to the nebulization efficiency. The nebulization efficiency is size dependent, with large (>500 nm) and small (<200−250 nm) BC particles nebulized with lower efficiency than mid-sized particles (Wendl et al., 2014;Schwarz et al., 2012). Besides, ice core samples were stored in the liquid phase after melting, and prior findings indicate that storage in the liquid phase can result in as great as 80% reduction in 530 measured BC concentrations (Wendl et al., 2014;Kaspari et al., 2014). Nevertheless, because we focus on the temporal variation of rBC from ice core for the historical records rather than on absolute concentrations, the systematic underestimation of rBC does not affect the historical changes.
Historical records from the East Rongbuk ice core showed that rBC concentrations have increased approximately threefold from 1975-2000 relative to 1860-1975 (APCC dataset I-5, andFig. 13), inferring anthropogenic BC is transported 535 to high elevation regions of the Himalaya (Kaspari et al., 2011). The Geladaindong ice core provided the first long-term rBC records in the central TP spanning 1843−1982; after the 1940s, the rBC record was also characterized by an increasing trend ( Fig. 13). Such an increase in rBC concentrations over the recent decades was likely attributed to the increased combustion emissions from regional BC sources, and a reduction in snow accumulation (Jenkins et al., 2016).

BC from lake sediment cores 540
The freeze-dried lake sediment core samples were grinded into powder with a size <0.074 mm using an agate mortar, and approximately 0.10−0.15 g of sample was transferred into a 50 mL centrifuge tube. In order to remove carbonates, silicates and metal oxides, the sediment samples were acid treated with 10 mL of HCl (2 N) and left for 24 hours at room temperature.
The supernatants were removed and rinsed with ultrapure water. Then, 15 mL of the mixture of HCl (6 N) and HF (48% v/v) with ratio of 1:2 (v/v) was added to the residue for further digestion for 24 h at room temperature and subsequently rinsed 545 thoroughly with ultrapure water. The residue was treated with HCl (4 N) at 60 °C overnight to get rid of fluoride that may have formed, which was then centrifuged to remove the supernatant liquid. The residue was rinsed with pure water until the pH of the eluent became neutral. Finally, the residual solid was diluted with 200 mL of ultrapure water and filtered through a 47 mm quartz fibre filter ensuring even distribution on the filter surface, which was then analysed for BC abundances (Cong et al., 2013). 550 Generally, the IMPROVE-A protocol with TOR has been used for determination of BC concentrations in lake sediments (Neupane et al., 2019). The repeated measurement of BC in a few lake sediments (n=5) were to ensure reproducibility of measurements, which reported as relative percentage deviation was better than 8%. Standard reference material (marine sediments, NIST SRM-1941b) were also analysed to assess the accuracy of measurements. It indicated an average accuracy of 5.5% for the measurements of BC in lake sediments in this study. 555 The BC flux (g m −2 yr −1 ) of the lake sediment cores was calculated by multiplying the BC concentration (mg g −1 ) with the quotient of total weight of sample (dry weight per layer) and area of gravity corer (6 cm inner diameter polycarbonate tube; area = π×r 2 ), as well as the deposition time (i.e., the time difference between the two consecutive deposition layers): Historical BC records from the lake sediment cores show an increasing trend of BC concurrent with increased 560 anthropogenic emissions patterns since the 1950s (APCC dataset I-5, and Fig. 14). The relatively constant trend of BC before the 1950s could be attributed to the background level with minimal inputs from anthropogenic activities (Cong et al., 2013;Neupane et al., 2019). Previous studies pointed out that the deposition of BC in lake sediment cores was mainly related to river transport from the lake basin as a result of climate change (e.g., increases in temperature and precipitation) (Li et al., 2017).
The much higher BC flux in recent decades may also be caused by increases in direct atmospheric deposition in addition to 565 riverine input.

Data availability
All the BC, OC, WIOC, WSOC, MAC values of BC and WSOC, and BC isotope datasets presented in this study have been released and are available for free download from the Cold and Arid Regions Science Data Center at Lanzhou (https://doi.org/10.12072/ncdc.NIEER.db0114.2021, Kang and Zhang, 2021). In each dataset, a short summary is also provided. Auxiliary data including site descriptions (e.g., locations), observation contents, and measurements are presented.

Conclusions
In this study, we present organic carbon and bThe dataset of BClack carbon and OC concentrations and their related MAC 580 values and carbon isotope signatures from the atmosphere, glaciers, snow cover, precipitation, and lake sediments from the TP and its surroundings based on the APCC program over the Third Pole region are presented in this paper based on the APCC program. These data are a collaborative effort aimed to address multiple scientific issues, including: (1) characterizing the carbonaceous components and depicting their spatial and temporal variations over the TP and its surrounding region; (2) identifying the carbon aerosol sources and investigating the mechanisms of long-range transport; and (3) constraining the role 585 of carbon components on the glacier/snow cover melting. Our continuous efforts based on the current APCC program are the in-depth investigations of the origins and distributions of atmospheric pollutants and their impacts and response to cryospheric changes over the Third Pole region. Moreover, this study paper presents the long-term spatial analysis of carbonaceous particles from multi environmental samples (atmosphere, glacier, lake sediments, etc.) over the Third Pole. The data from this work are also significant for understanding the negative consequences of pollution on environment of remote regions and pave a way 590 for future perspectives and protection strategies.

Competing interests
The authors declare that they have no conflict of interests.

Figure 13.
Historical BC concentration records retrieved from ice cores over the Himalayas and Tibetan Plateau. East Rongbuk ice core data from the northern slope of Mt. Everest in Himalayas (Kaspari et al., 2011), and Geladaindong ice core data from the Tanggula Mountains of the Central central TP Tibetan Plateau (Jenkins et al., 2016).

Figure 14.
Historical BC flux (g m −2 yr −1 ) retrieved from lake sediment cores across the Third Pole. Nam Co lake sediment cores data was cited from (Cong et al., 2013); other lake sediment cores data was cited from (Neupane et al., 2019).