1. Introduction
Due to the high sensitivity to human activities and climate change (Bibi et al. 2018), a wetter Tibetan Plateau (TP) against the backdrop of global warming has received extensive worldwide concerns during recent decades (He et al. 2021; Tian et al. 2001; Wang et al. 2008; Yang et al. 2011; Yu et al. 2022). On one hand, the spatiotemporal variations of precipitation directly regulate the recharging of TP (Wu et al. 2014; Zhang et al. 2013; G. Zhang et al. 2020), and cause extreme events in downstream regions. On the other hand, the latent heating released from precipitation plays a crucial role in the local heating source intensity over the TP, subsequently inducing large-scale circulation anomalies and influencing weather and climate (Wu et al. 2016).
Multisource reanalysis datasets and observations have demonstrated that the TP has become wetter in recent decades (Qiu 2008; Zhou et al. 2019). Among all seasons, summer precipitation not only contributes the largest portion to the annual mean but also shows the most intense changes (Ma et al. 2016). Moreover, the spatial distribution of summer precipitation, under the influence of climate warming, shows significant heterogeneity (Liu et al. 2021b; Zhou et al. 2019). Previous studies based on in situ data have indicated an increase in summer precipitation over the northern TP (Zhang et al. 2022) while the southern TP experiences the opposite trend (Liu et al. 2021a,b). Additionally, reanalysis datasets have revealed a substantial increase in precipitation over the central-western TP since the mid-1990s, contributing significantly to the overall wetting of the entire TP (Bibi et al. 2018; Sun et al. 2020). Besides that, a significant decrease in summer precipitation has been observed over the southeastern TP (Kitoh and Arakawa 2016; Yang et al. 2011, 2014). The variability of summer water vapor also confirms the inhomogeneous changes (He et al. 2021; Yu et al. 2022).
Given the significance of spatial diversity in summer precipitation for the local environment and climate, numerous studies have investigated the underlying mechanisms. On the interdecadal scale, there is an inverse relationship between summer precipitation over the southern and northern TP. The decrease in precipitation over the southern TP since the late 1990s has been attributed to a weakened sensible heat source, whereas the opposite trend over the northern TP has been linked to the influences of the interdecadal Pacific oscillation and Atlantic multidecadal oscillation (AMO) on the Asian westerly jet and Rossby waves (Liu et al. 2021b). Another study has connected this dipole mode to the wave propagation anomalies induced by the summer North Atlantic Oscillation (NAO) (Liu et al. 2021a). Moreover, a warming Arabian Sea has been found to promote precipitation in the northern TP by inducing westward movement of the Indian summer monsoon and a poleward shift of summer monsoon over the TP (Zhang et al. 2022). It should be noted that the abnormal atmospheric circulation configuration is characterized by an anomalous anticyclone (cyclone) on the eastern (western) TP. Similar circulation patterns have been observed in studies investigating the significant anomalies of summer precipitation over the central-western TP, attributed to the influences of the AMO on the subtropical westerly jet stream (Sun et al. 2020). Further, the contribution from the Indian summer monsoon has also been suggested through the explorations of water vapor transport from the Indian Ocean, which tends to move westward and contributes to the increased precipitation over the western TP (He et al. 2021).
In the above context, most studies have focused on elucidating the mechanisms behind nonuniform variations in summer precipitation by emphasizing the contributions from the low and midlatitudes, such as the Arabian Sea (Zhang et al. 2022). This raises the question of whether there is a connection between these nonuniform variations and high-latitude forces. It is widely recognized that the Arctic climate is experiencing unprecedented changes due to global warming. The role of sea ice has been discussed worldwide because of its significant impact on the climate of the Northern Hemisphere (Gu et al. 2018). These impacts include the influence on extreme weather in boreal midlatitudes through weakened zonal winds and increased wave amplitudes (Deng et al. 2020; Francis and Vavrus 2012), the modulation of Indian summer monsoon rainfall extremes (Chatterjee et al. 2021), the variability of the East Asian summer monsoon through anomalous Eurasian wave trains (Guo et al. 2013; Wu et al. 2009), the summer precipitation patterns over East Asia by exciting Silk Road pattern (SRP) anomalies (He et al. 2017), and the spatially consistent variations in surface air temperature over central Asia through the induction of abnormal diabatic heating over the Greenland Sea (Qin et al. 2022). Most of these studies have highlighted the significance of the Rossby waves and westerlies over Eurasia. Additionally, the importance of memory effects associated with land processes in Eurasia has been demonstrated through a series of general circulation model experiments (Nakamura et al. 2019). The combined effects of declining Arctic sea ice and mid-/high-latitude Eurasian snow cover fraction not only influence European heat waves by inducing drier soil and stronger heat flux (R. Zhang et al. 2020) but also serve as complementary precursors for Chinese summer rainfall variability (Wu et al. 2009). Moreover, Li et al. (2018) revealed that the preceding sea ice loss in the Barents Sea enhances summer hot drought events by regulating snow depth and soil moisture in Eurasia.
Numerous studies have investigated the connections between the TP and tropical climate (He et al. 2021; Zhang et al. 2022), suggesting that the TP may serve as a bridge linking Arctic and tropical anomalies. However, there is still a need to further understand the importance of connecting climate change on the TP with rapid variations in Arctic sea ice. Currently, there are limited studies on this topic. For instance, it has been found that Barents–Kara sea ice loss could intensify an anomalous tripole mode in air temperature over the Northern Hemisphere, leading to enhanced winter warming on the TP (Duan et al. 2022). Furthermore, positive anomalies of winter Barents Sea ice on the interannual scale have been closely linked to negative snow depth anomalies in the midwestern TP through westerlies and Rossby waves (Chen et al. 2020b), as well as the NAO (Chen et al. 2020a). However, the specific mechanisms linking Arctic sea ice changes with summer precipitation anomalies over the TP remain unclear. Further research is needed to explore this relationship in more detail.
This study aims to examine whether the Arctic sea ice regulates the inhomogeneous summer precipitation over the TP and to explore if there are distinct mechanisms involved from late spring to early summer. The data and methods employed in this study are described in section 2. The connections between Arctic sea ice and the uneven summer precipitation patterns over the TP are discussed in section 3, which includes the presentation of simulation results. Last, a summary and further discussion are provided in section 4.
2. Data and methods
a. Data sources
The in situ daily accumulative precipitation observations from the China Meteorological Administration are used. The dataset comprises data from 113 stations that have maintained continuous observations since 1961, as depicted in Fig. 1. These observations have undergone a rigorous multistep quality control process and are considered to provide the most reliable representation of precipitation characteristics over the TP (Liu et al. 2021b). Meantime, limited by the lack of in situ dense observational data, particularly for data in the central-western areas (Fig. 1a), reanalysis datasets are widely used in the research about the TP to overcome the difficulty. The monthly precipitation data produced by the National Oceanic and Atmosphere Administration (NOAA) (https://psl.noaa.gov/data/gridded/data.precl.html) is applied, which have a horizontal resolution as 1.0° × 1.0° and cover from 1948 to the present. This dataset is defined by interpolation of gauge observations and has shown good relationships with those in several published gauge-based datasets (Chen et al. 2002). Furthermore, comparisons conducted among multiresource precipitation datasets have confirmed the suitability and reliability of the NOAA data for the TP (see details in the online supplemental material).
Spatial distributions of the July precipitation during 1961–2020. (a) Climatic precipitation (unit: mm day−1) and (b) linear trends of standardized precipitation (unit: 10−2 mm day−1 yr−1) based on in situ observations. (c),(d) As in (a) and (b), respectively, but for NOAA dataset. (e) The correlations and (f) the root-mean-square error (RMSE; unit: mm day−1) between the observational dataset and NOAA dataset. The two magenta parallel lines in (b) and (d) delimit the geographic scope of the selected stations. The coefficients passing the significance test at the 0.05 level are presented by solid dots in (b) and (e) and by stippled areas in (d).
Citation: Journal of Climate 36, 22; 10.1175/JCLI-D-23-0027.1
In addition, the ERA5 dataset, developed by the European Center for Medium-Range Weather Forecasts (Hersbach et al. 2020), has emerged as a state-of-the-art reanalysis dataset widely used in recent studies. Previous research has demonstrated that this dataset exhibits superior suitability for performing spatial maps of annual variances in water vapor over the TP compared to other reanalysis datasets (He et al. 2021; Zhao and Zhou 2019). Obtained from the ERA5, the monthly wind variables, geopotential height, specific humidity, and temperature fields are used to analyze changes in atmospheric circulations, water vapor flux (WVF), and Eady growth rate (EGR). Furthermore, the monthly means of daily forecast accumulated snowmelt from ERA5 are used to assess their contributions to soil moisture changes. The ERA5 dataset employed in this study has a resolution of 0.5° × 0.5° horizontally and consists of 37 vertical levels.
Long-term monthly sea temperature and sea ice concentration are sourced from the Met Office Hadley Centre (Rayner et al. 2003), with a horizontal resolution of 1.0° × 1.0° and 2.5° × 2.5°, respectively. Additionally, monthly soil moisture data covering the period from 1961 to the present are obtained from the Global Land Data Assimilation System (Rodell et al. 2004). The soil moisture data used in this study have a horizontal resolution of 0.25° and specifically focus on the 10–40-cm layer to capture more reliable memory effects. To ensure consistency among multiple datasets, the climatological variables used in this study are calculated based on the period spanning from 1961 to 2020.
b. Methods
Moisture transportation plays a direct and crucial role in precipitation changes (Ma et al. 2020), and therefore, it is essential to delve into this in detail. The vertical integrated WVF between two specific layers is calculated using the method proposed by Trenberth (1991). To capture the moisture transportation around the TP and its surrounding areas, the WVF is integrated between 600 and 400 hPa, considering the altitude. Meanwhile, the horizontal wave activity flux (WAF) developed by Takaya and Nakamura (2001) is employed to diagnose the wave propagation induced by relevant factors. The EGR at 700 hPa is calculated to determine atmospheric baroclinic responses to changes in the underlying surface (Du et al. 2022; Lindzen and Farrell 1980). Other methods utilized in this study include linear regression and Pearson correlation analyses. The significance is tested using Student’s t test and results would be statistically significant at 0.10 level at least.
The SRP is defined as a wavelike train that propagates eastward along the Asian westerly jet. It is identified as the leading empirical orthogonal function mode of the 200-hPa meridional wind anomalies within the domain 30.0°–50.0°N, 0.0°–150.0°E (Enomoto et al. 2003; Hong et al. 2022; Yasui and Watanabe 2010). The first principal component of the SRP mode is extracted and standardized, and it is referred to as the SRP index (SRPI). The SRPI serves as a representative indicator of the variability associated with the SRP pattern.
The Community Earth System Model (version 1.0.4), developed by the National Center for Atmospheric Research, consists of models representing the atmosphere, ocean, land, and sea ice (Hurrell et al. 2013). To provide evidence supporting the impacts of sea ice loss on precipitation anomalies over the TP, the Community Atmospheric Model version 5.1 (CAM5) is employed. CAM5 has a finite-volume grid with a resolution of 1.9° × 2.5° and 30 hybrid sigma–pressure levels. For the sensitivity experiment, the boundary conditions are set by prescribing the sea ice concentration and sea surface temperature (SST), while all other external variables are fixed. The control run is conducted for 50 years, driven by climatological monthly means of sea ice and SST. As for the sensitivity experiment, the linear decrease of Arctic sea ice concentration during 1961–2020 is calculated for each grid. And the boundary conditions of sea ice concentration over key regions are set as the boundary condition in the control experiment plus the linear decreases in the corresponding months. Meanwhile, the boundary condition of SST is forced as −1.8°C where sea ice concentration equals zero.
3. Results
a. Changes in summer precipitation and their relations with Arctic sea ice concentration
The spatial distribution of precipitation over the TP exhibits variations across different months. Among the summer months, July is considered to be more representative, and variations in July precipitation show closer relationships with anomalous Arctic sea ice (additional details can be found in the supplemental material). Therefore, the subsequent discussions in this study specifically focus on July precipitation over the TP. Figure 1 shows that the NOAA precipitation data have good agreements with the in situ data from 1961 to 2020. The data not only exhibit high correlation coefficients for most stations on the TP (greater than 0.50), significant at the 95% confidence level, but also demonstrate relatively low root-mean-square errors, mostly less than 1 mm day−1. The results reveal that the NOAA dataset can capture the characteristics of July precipitation over the TP and is employed in the following analysis.
The climatological pattern in July precipitation over the TP dramatically decreases from the southeastern to the northwestern regions, as evidenced by both the in situ dataset and reanalysis data. This pattern is consistent with previous studies (Feng and Zhou 2012; He et al. 2021; Xu et al. 2020). The minimum precipitation is generally observed in the Qaidam Basin and the western Himalayas. However, the standardized precipitation during 1961–2020 exhibits significant increasing trends along the northern, hinterland, and southwestern regions of the TP. These anomalous increases, captured by both reanalysis and observation, suggest a northward and westward extension of July precipitation regions. This abnormal pattern has been discussed by Zhang et al. (2022), who explored its relationship with the northward extension of the summer monsoon over the TP. However, it remains unclear whether there are potential linkages between this abnormal precipitation pattern and decreased Arctic sea ice concentration. While anomalies derived from reanalysis data may provide more intuitive and evident results, the results are supposed to be more realistic using in situ precipitation. In Fig. 1, stations located within two magenta parallel lines are selected to highlight the heterogeneous variations in precipitation, and the standardized daily mean July precipitation at these stations is defined as PI.
Previous studies have indicated that the responses of atmospheric circulation to changes in Arctic sea ice are nonlinear (Chen et al. 2016; Screen 2017). Therefore, it is crucial to investigate the influences of sea ice variations during different study periods. To explore the possible lagged and contemporaneous impacts, the spatial correlations between Arctic sea ice concentration from April to July and the PI are calculated separately for each month. The results depicted in Fig. 2 suggest that the monthly changes in Arctic sea ice during late spring [April and May (AM)] and early summer [June and July (JJ)] are potentially associated with the anomalous precipitation pattern that extends to the northern and western TP.
Spatial correlations between PI and Arctic sea ice concentration in (a) April, (b) May, (d) June, and (e) July. The coefficients passing the significance test at the 0.05 level are presented by stippled areas. The magenta rings represent the selected regions for the SICI. Also shown are time series of PI and SICI during (c) April and May and (f) June and July. The PI is denoted by bars, and solid lines present SICI. Correlations between PI and SICI are listed in the lower left; triple asterisks (***) indicate coefficients significant above the 99% level.
Citation: Journal of Climate 36, 22; 10.1175/JCLI-D-23-0027.1
As for the preceding months (i.e., AM), there is a statistically significant correlation zone that covers the region from the east Greenland coast to the Kara Sea (65.0°–82.0°N, 30.0°W–80.0°E). Taking this region as the key study zone, the corresponding spatial-weighted domain-averaged sea ice concentration is calculated and defined as SICI during April and May. The changes in both SICI and PI are opposite clearly, with correlation coefficients smaller than −0.50, significant at the 99% confidence interval. The spatial correlations between different seasons exhibit evident differences. Significant negative correlations in JJ are observed in the region spanning from the east Greenland coast to the Laptev Sea (65.0°–83.0°N, 30.0°W–150.0°E), which extend farther east compared to those in AM and are consistent with the seasonal variations of Arctic sea ice (figure not shown). The domain-averaged sea ice concentration (after spatial weighting) over this zone is then defined as SICI in JJ. There is a strong relationship between SICI during June/July and PI, with correlation coefficients of −0.51 and −0.53, respectively (above 99% confidence level).
To represent the reduced Arctic sea ice during AM and JJ over respective key regions, the temporally averaged SICI during AM and JJ are denoted as the SICI(AM) and SICI(JJ), respectively. The SICI(AM) and SICI(JJ) exhibit correlation coefficients with PI at −0.52 and −0.53, respectively, both passing the significance test at the 0.01 level. The good relationships raise the question of the potential physical processes linking the reduction in preceding and current sea ice with the July precipitation pattern.
b. Physical associations between precipitation anomalies and the late-spring sea ice loss
The results mentioned above indicate the impacts of sea ice loss during AM on the July precipitation anomaly. However, it is unclear how these connections work across seasons. In general, the land processes have longer memory compared to the atmosphere, which could affect the atmosphere on a time scale beyond a season (Matsumura and Yamazaki 2012). Previous studies have emphasized the role of memorized anomalies in affecting the atmosphere and amplifying the atmospheric responses associated with Arctic sea ice loss (Nakamura et al. 2019). Given that soil moisture anomalies could persist for 2–3 months and significantly influence climate changes through land–atmosphere coupling, Fig. 3 focuses on soil moisture from AM to July and presents regression maps against SICI(AM), and exhibits spatial correlations with PI as well. Notably, distinguishing from most regions across the Eurasian continent, the soil moisture anomalies over the eastern Caspian Sea and northeastern TP not only last for months but are also closely correlated with SICI(AM) and PI simultaneously.
Regression fields of soil moisture (m3 m−3) in (a) AM, (c) June, and (e) July against the inverted SICI(AM). Correlation maps between soil moisture in (b) AM, (d) June and (f) July and PI. The coefficients passing the significance test at the 0.05 level are presented by stippled areas. The magenta rectangular boxes represent the selected regions for the SMI. Time series of (g) SMI and SICI(AM) and (h) SMI and PI. The SICI(AM) and PI are denoted by bars in (g) and (h), respectively, and solid lines present SMI. Correlations are listed in the lower left, the single, double, and triple asterisks indicate coefficients significant above the 90%, 95%, and 99% levels, respectively.
Citation: Journal of Climate 36, 22; 10.1175/JCLI-D-23-0027.1
The SMI index is defined based on the differences between spatially averaged soil moisture over the eastern Caspian Sea (40.0°–52.5°N, 52.5°–75.0°E) and northeastern TP (42.5°–55.0°N, 90.0°–115.0°E). The time series of SMI from AM to July show highly synchronized variations and exhibit strong opposite changes with SICI(AM), with correlation coefficients of approximately −0.40, significant at the 99% confidence level. Meanwhile, Fig. 3h displays closely positive relationships between SMI and PI, with coefficients near 0.30, passing the significance test at the 0.10 level at least. In detail, the reduction of sea ice during AM leads to abnormally wet soil (dry soil) in the eastern Caspian Sea (northeastern TP). The soil anomalies might play an important role in linking the heterogeneous precipitation pattern over TP. The following would explore the robustness of the role of soil moisture in precipitation variations.
1) Connections between sea ice loss and anomalous soil moisture
As discussed by many researchers (Chen et al. 2020b; Duan et al. 2022; Li et al. 2018), the decline of Arctic sea ice brings increased turbulent heat flux and stimulates a Rossby wave that spreads horizontally and connects the climate over the Arctic and Eurasia. Sea ice variations are also closely related to snow depth/snow cover over the Eurasian continent, which further influences soil conditions (Li et al. 2018; R. Zhang et al. 2020). Meanwhile, boreal snowmelt in May is particularly crucial for soil moisture anomalies, as they could persist into early summer and might be more pronounced due to changes in solar radiation changes. Figure 4 focuses on these relationships and examines the potential linkages between soil anomalies and the reduction of sea ice in late spring.
Regressions of 700-hPa EGR (shading; units: 10−5 s−1) and 300-hPa horizontal WAF (arrows; units: m2 s−2) against the concurrent inverted SICI in (a) April and (b) May. The composed energy weaker than 0.050 and 0.035 (arrows; units: m2 s−2) is not displayed in (a) and (b), respectively, for better presentation. Regressed temperature advection (shading; units: 10−6 K s−1) and horizontal winds (arrows; units: m s−1) at 500 hPa against the inverted SICI in (c) April and (d) May. (e) Regressed snowmelt (units: 10−4 m) in May against inverted SICI during AM. The magenta rectangular boxes represent the selected regions for the SMI and the green lines depict the wave trains. The coefficients passing the significance test at the 0.10 level are presented by stippled areas and blue arrows.
Citation: Journal of Climate 36, 22; 10.1175/JCLI-D-23-0027.1
Figure 4a presents an anomalous wave train that disperses from the east of the Greenland coast to the eastern Caspian Sea in April, propagating along the jet axis in the midlatitudes but relatively weaker in spreading to the northeastern TP. As indicated by the divergence of WAF, reduced sea ice concentration over the eastern Greenland coast is important in driving this anomalous wave train. Besides the impacts of associated heat flux (Deser et al. 2000; Germe et al. 2011), which have been verified through the linear baroclinic mode (Qin et al. 2022), sea ice loss in this region is often accompanied by SST anomalies in the North Atlantic (Deng and Dai 2022). The anomalous SST pattern could change the regional temperature gradient (Du et al. 2022), resulting in anomalies in cyclonic vorticity and baroclinicity structures and further contributing to the formation of the midlatitudinal wave train (Peng et al. 2003; Qin et al. 2022; Sutton et al. 2000). The positive anomalies in baroclinicity are consistent with the energy dispersion shown in Fig. 4a, suggesting the potential impacts of sea ice loss in April on the abnormal climate over the eastern Caspian Sea. As Arctic sea ice melting progresses into May, the regressed energy in May disperses through two paths. One path propagates from the subpolar Atlantic, turns southeast around the Barents–Kara Seas, and primarily affects the northeastern TP. The other path spreads from the high-latitude Atlantic to the eastern Caspian Sea via the East European Plain.
As a result of energy sinking, there is a positive anomaly in geopotential height over the eastern Caspian Sea during AM (figures not shown), which facilitates downward motion and adiabatic warming. This, in turn, leads to increased snow melting and contributes to wet soil through hydrological effects. Conversely, the negative anomaly over the northeastern TP during May shows lower soil moisture. Temperature advection plays a significant role in these processes, with southwesterly anomalies and warm advection toward the eastern Caspian Sea in both April and May. These anomalies result in warming and contribute to snowmelt in May. However, the easterly anomalies toward the northeastern TP are accompanied by an anticyclone over the high-latitude landmass, which tends to transport cold air masses. As a result, there is significant cold advection toward the northeastern TP, which hinders further snow melting and leads to dry soil conditions. It should be noted that as spring progresses, the climatic snow-cover area retreats northward (figure not shown), and thus inhomogeneous distribution of snowmelt over the eastern Caspian Sea is reasonable. Excessive snowmelt could also impact the recharge of rivers, such as the Sil River and the Amu Darya, which might further affect downstream soil moisture.
2) Linkages of anomalous soil moisture to uneven July precipitation on TP
Much research has demonstrated that SRP changes are closely related to the summer precipitation on TP. As well known that the SRP is characterized by alternating positive and negative meridional wind anomalies over Eurasia, spreading horizontally from the Asian jet entrance, particularly the Caspian Sea, to East Asia. The thermal conditions over the Caspian Sea are crucial for the excitation of the SRP (Hong et al. 2022). The abnormal dipole-like soil conditions resulting from the reduction of sea ice during AM might act as a role in regulating the intensity and phase of the SRP, which could contribute to the extension of July precipitation over TP to the north and west.
According to Fig. 3, the anomalous soil moisture over the selected regions during AM persists until JJ and Fig. 5 further explores the potential connections between anomalous soil moisture and the SRP in July. The SMI during JJ exhibits a strong correlation with the SRPI, with a correlation coefficient of 0.36 (above 99% confidence level). Meanwhile, the regressed 200-hPa meridional winds based on SMI(JJ) and SRPI show similar anomalous centers, with a high spatial correlation of 0.90 (based on the same region used for SRP definition), passing the significance test at the 0.01 level. The regressed 300-hPa WAF (figure not shown) against SMI(JJ) is consistent with this against SRPI, which further validates their close relationship. It is evident that the abnormal soil moisture strengthens the eastward propagation of the SRP and favors the deepening of the wave trough and amplification of the ridge to the west and east of TP, respectively. The enhanced wave amplitude supports the formation of a dipole mode around the TP, characterized by a cyclone in the western TP and an anticyclone in the eastern TP. Such dipole anomalies, as also identified by Zhang et al. (2022), facilitate strengthened moisture transportation to the northern and western TP, ultimately contributing to the northward and westward extension of July precipitation.
Regression maps of (a) meridional winds in 200 hPa (units: m s−1) against SMI in JJ (shaded) and SRPI in July (contours; interval: 1). The coefficients passing the significance test at the 0.05 level are presented by stippled areas and purple contours. The solid and dashed contours indicate positive and negative values, respectively. The green dash–dotted line depicts the predominant wave train at 300 hPa regressed by SRPI. (b) Time series of SMI in JJ and SRPI in July from 1961 to 2020. Correlation is listed in the upper left and exceeds 0.01 confidence level as denoted by the triple asterisks.
Citation: Journal of Climate 36, 22; 10.1175/JCLI-D-23-0027.1
The first column of Fig. 6 illustrates the effects of SMI on the abnormal July precipitation pattern over the TP. The regressions of 500-hPa geopotential height and wind fields against SMI(JJ) present a dipole pattern, characterized by a negative center over the western TP and a positive center over the northeastern TP. These anomalies are accompanied by southwest flows on the west of the TP, enhancing the input of water vapor to the western TP. Meanwhile, the abnormal easterly winds associated with the anticyclone inhibit the output of water vapor from the TP, leading to moisture accumulation over the northern TP. As a result, significant moisture convergence is observed in the western and northern TP regions, as shown in Fig. 6c. This moisture convergence corresponds to significant positive anomalies in July precipitation, as depicted in Fig. 6e.
Regression fields of geopotential height (shaded; units: gpm) and horizontal winds (arrows; units: m s−1) in July against (a) SMI and (b) inverted SICI in JJ. The green dash–dotted line depicts the Rossby wave train in 300 hPa regressed by inverted SICI(JJ). (c),(d) As in (a) and (b), respectively, but for the vertically integrated divergence of moisture flux (shaded; units: 10−8 kg m−2 s−1) and WVF (arrows; units: 10−2 kg m−1 s−1) between 600 and 400 hPa. (e),(f) As in (a) and (b), respectively, but for July precipitation (shaded; units: mm day−1). The coefficients passing the significance test at the 0.05 level are presented by stippled areas and blue arrows.
Citation: Journal of Climate 36, 22; 10.1175/JCLI-D-23-0027.1
c. Impacts of early-summer sea ice loss on precipitation anomalies
The strong correlation between SICI(JJ) and PI implies a potential connection between Arctic sea ice reduction and the inhomogeneous precipitation changes over the TP in July. Compared to the cross-seasonal impacts, early-summer sea ice loss influences are more direct. As illustrated in Fig. 6b, the regression of 500-hPa geopotential height exhibits the anomalous dipole mode as well, with the center of the west cyclone located relatively southward. This dipole pattern in geopotential height corresponds to the moisture convergence flux regressed by the inverted SICI(JJ). The enhanced southwestward moisture input and weakened eastward water vapor output indicate increased water vapor over the western and northern TP, which facilitate the northward and westward extension of July precipitation over the TP. Furthermore, the regressed 300-hPa WAF (Fig. 6b) suggests that the loss of sea ice in JJ enhances the propagation of a wave train from the Arctic through Eurasia and eventually reaches the TP, with energy sinking over the western and northeastern TP (figure not shown). This further supports the connection between early-summer sea ice loss and the spatial distribution of July precipitation anomalies over the TP.
d. Model and experiment
To assess the impacts of sea ice loss during AM and JJ on the anomalous increases in July precipitation over the northern and western TP, three numerical experiments (including a control experiment and two sensitivity experiments) are conducted. The sensitivity experiments related to sea ice loss during AM (case_AM) and JJ (case_JJ) are repeated 50 times with distinct initial conditions from control runs, integrated from 1 April to 31 July and from 1 June to 31 July, respectively. The detailed experimental design can be seen in section 2. After approximately 15 model years, the simulations reach a quasi-equilibrium state where the differences in July precipitation between the sensitivity experiments and the control runs become stable. Focusing on the key region over the TP as shown in Fig. 1, the July precipitation differences between the sensitivity experiments and the control experiment from the 16th to the 50th year are displayed in Fig. 7. The results show that anomalies of July precipitation are positive in most years and range within 3 (2) mm day−1, with an average of 1.04 (0.99) mm day−1 for case_AM (case_JJ). The simulations provide evidence for the lagged and current impacts of reduced Arctic sea ice on atmospheric circulations and moisture transportation around the TP in July, which subsequently induce the extension of TP precipitation to the north and west.
CESM-simulated precipitation anomalies (units: mm day−1) during July for the last 35 model years (years 16–50), related to (a) case_AM and (b) case_JJ.
Citation: Journal of Climate 36, 22; 10.1175/JCLI-D-23-0027.1
Reduced Arctic sea ice concentration during AM excites energy that disperses southeast and affects soil moisture over the eastern Caspian Sea and northeastern TP, further regulating the SRP and driving the dipole mode. Figures 8a, 8c, and 8e display anomalies in atmospheric circulations, water vapor transportation, and July precipitation, respectively, associated with case_AM. The dipole mode around the TP is evident, with a deepened wave trough over the western TP and an amplified wave ridge on the northeastern TP, although the anomalous anticyclone over the northeastern TP is weaker compared to the results in section 3b. Additionally, the cyclonic anomaly to the west of TP shifts slightly eastward. The cyclonic anomaly over the western TP strengthens more moisture transported to the western TP and the anomalous anticyclone over the northeastern TP is unhelpful for the water outputting from the east margin. Accompanied by these responses, the positive precipitation anomalies over the northern and western TP are evident, although with a significant area located relatively east.
Model-simulated differences of (a) 500-hPa geopotential height (shaded; units: 10 gpm) and horizontal winds (arrows; units: m s−1); (c) vertically integrated divergence of moisture flux (shaded; units: 10−7 kg m−2 s−1) and WVF (arrows; units: 10−1 kg m−1 s−1) between 600 and 400 hPa; (e) July precipitation (shaded; units: mm day−1) between case_AM and control experiments. (b),(d),(f) As in (a), (c), and (e), respectively, but for differences between case_JJ and control runs. The values passing the significance test at the 0.10 level are presented by stippled areas and blue arrows.
Citation: Journal of Climate 36, 22; 10.1175/JCLI-D-23-0027.1
As for the model-simulated differences related to case_JJ, the energy divergence induced by Arctic sea ice loss in JJ spreads from the subpolar jet to the subtropical jet, resulting in a dipole mode of 500-hPa geopotential height around the TP. Compared to the diagnostic results, the anticyclone over the northeastern TP is weaker but significantly inhibits the water output from the eastern TP. On the other hand, the cyclone over the western TP is stronger. The forward-tilting trough is favorable for blocking activities and amplified wave structures around the TP, enhances water vapor input to the western TP, and promotes moisture convergence in the western and northern TP. These changes in atmospheric circulations are associated with increased July precipitation over the TP, as depicted in Figs. 8d and 8f.
4. Conclusions and discussion
As global warming occurs, the July precipitation over the TP and Arctic sea ice both present dramatic variations. The former has shown a tendency to extend northward and westward and the latter has experienced tremendous loss, which is known to be related to climate changes in midlatitudes. This study aims to investigate the lagged and concurrent relations between them and explore the underlying mechanisms.
The results indicate strong connections between Arctic sea ice loss from April to July and anomalous July precipitation over the TP, including two distinct pathways. First, the reduced sea ice in late spring (April and May) excites anomalous wave trains that propagate to midlatitudes, leading to positive (negative) anomalies in 500-hPa geopotential height and significant warm (cold) advection over the eastern Caspian Sea (northeastern TP). The anomalous advections lead to changes in snow melting, which impacts soil moisture through hydrological processes. Due to the memory effects of soil moisture, this dipole-like anomaly during late spring could maintain to July, strengthens the midlatitude Silk Road pattern, and brings the dipole mode around TP, characterized by a west cyclone and an east anticyclone. Second, the decline in sea ice during early summer (June and July) directly induces an anomalous Rossby wave that spreads from the Arctic to the TP via Eurasia. Responses in 500-hPa geopotential height field and moisture transportation between 600 and 400 hPa show a similar dipole mode around the TP. Consequently, the circulation configurations contribute to enhanced southwesterly winds that transport warm and humid moisture from the south to the western TP, while easterly anomalies inhibit the climatological westerly winds and prevent water vapor from outputting the northern TP. The combined effects lead to moisture convergence in the western and northern TP, ultimately increasing precipitation. The mechanisms are illustrated in Fig. 9.
Mechanisms connecting the reduced Arctic sea ice with July precipitation changes over TP by modulating land processes and Rossby waves.
Citation: Journal of Climate 36, 22; 10.1175/JCLI-D-23-0027.1
Further comprehensive investigations are needed to deepen our understanding of the impacts of Arctic sea ice loss on uneven summer precipitation variations over the TP. First, it is crucial to assess the degree to which Arctic sea ice loss contributes to summer precipitation anomalies, considering the influences of other low-/midlatitude forces such as the warming Arabian Sea (Zhang et al. 2022), the AMO and NAO (Liu et al. 2021a,b), and heating conditions over the TP itself (He et al. 2021). Understanding the relative contributions of these factors will provide insights into the individual and combined effects on TP precipitation. Second, the potential combined effects between sea ice changes and other forces should be explored. Studies have suggested that anomalous sea ice could modulate the snow cover over the TP (Chen et al. 2020a,b), influencing the thermal conditions through hydrological effects (Zhu et al. 2008) and further impacting summer precipitation (He et al. 2021). Additionally, the AMO and NAO might be the bridges, considering they have been verified connected with Arctic sea ice loss through troposphere–stratosphere interactions (Screen 2017) and variations in baroclinic structures over sea ice melting regions (Du et al. 2022). Furthermore, previous studies have identified interdecadal changes in TP precipitation around the late 1990s and early 2000s (Liu et al. 2021b; Sun et al. 2020; Zhang et al. 2022). Thus it would be valuable to assess whether there are links between July precipitation variations and Arctic sea ice loss on different time scales.
Acknowledgments.
This research is supported jointly by the National Natural Science Foundation of China (Grants 42088101), the Second Tibetan Plateau Scientific Expedition and Research (STEP) program (2019QZKK0105), and the National Natural Science Foundation of China (Grant 41975083).
Data availability statement.
The monthly NOAA precipitation data at 1.0° × 1.0° resolution are derived from https://psl.noaa.gov/data/gridded/data.precl.html. The ERA5 dataset is available at https://cds.climate.copernicus.eu/ and has a resolution of 0.5° × 0.5° horizontally and 37 levels in the vertical direction. Long-term monthly sea temperature and sea ice concentration are from https://www.metoffice.gov.uk/hadobs/hadisst/, with a horizontal resolution of 1.0° × 1.0° and 2.5° × 2.5°, respectively. The monthly soil moisture at 0.25° × 0.25° is from https://earth.gsfc.nasa.gov/hydro/data/gldas-global-land-data-assimilation-system-data. The in situ daily accumulative precipitation observations are provided by the China Meteorological Administration.
REFERENCES
Bibi, S., L. Wang, X. Li, J. Zhou, D. Chen, and T. Yao, 2018: Climatic and associated cryospheric, biospheric, and hydrological changes on the Tibetan Plateau: A review. Int. J. Climatol., 38 (Suppl. 1), e1–e17, https://doi.org/10.1002/joc.5411.
Chatterjee, S., M. Ravichandran, N. Murukesh, R. P. Raj, and O. M. Johannessen, 2021: A possible relation between Arctic sea ice and late season Indian summer monsoon rainfall extremes. npj Climate Atmos. Sci., 4, 36, https://doi.org/10.1038/s41612-021-00191-w.
Chen, H. W., F. Zhang, and R. B. Alley, 2016: The robustness of midlatitude weather pattern changes due to Arctic sea ice loss. J. Climate, 29, 7831–7849, https://doi.org/10.1175/JCLI-D-16-0167.1.
Chen, M., P. Xie, J. E. Janowiak, and P. A. Arkin, 2002: Global land precipitation: A 50-yr monthly analysis based on gauge observations. J. Hydrometeor., 3, 249–266, https://doi.org/10.1175/1525-7541(2002)003<0249:GLPAYM>2.0.CO;2.
Chen, Y., A. Duan, and D. Li, 2020a: Connection between winter Arctic sea ice and west Tibetan Plateau snow depth through the NAO. Int. J. Climatol., 41, 846–861, https://doi.org/10.1002/joc.6676.
Chen, Y., A. Duan, and D. Li, 2020b: Atmospheric bridge connecting the Barents Sea ice and snow depth in the mid-west Tibetan Plateau. Front. Earth Sci., 8, 265, https://doi.org/10.3389/feart.2020.00265.
Deng, J., and A. Dai, 2022: Sea ice-air interactions amplify multidecadal variability in the North Atlantic and Arctic region. Nat. Commun., 13, 2100, https://doi.org/10.1038/s41467-022-29810-7.
Deng, K., X. Jiang, C. Hu, and D. Chen, 2020: More frequent summer heat waves in southwestern China linked to the recent declining of Arctic sea ice. Environ. Res. Lett., 15, 074011, https://doi.org/10.1088/1748-9326/ab8335.
Deser, C., J. E. Walsh, and M. S. Timlin, 2000: Arctic sea ice variability in the context of recent atmospheric circulation trends. J. Climate, 13, 617–633, https://doi.org/10.1175/1520-0442(2000)013<0617:ASIVIT>2.0.CO;2.
Du, Y., J. Zhang, S. Zhao, and Z. Chen, 2022: A mechanism of spring Barents Sea ice effect on the extreme summer droughts in northeastern China. Climate Dyn., 58, 1033–1048, https://doi.org/10.1007/s00382-021-05949-9.
Duan, A., and Coauthors, 2022: Sea ice loss of the Barents-Kara Sea enhances the winter warming over the Tibetan Plateau. npj Climate Atmos. Sci., 5, 26, https://doi.org/10.1038/s41612-022-00245-7.
Enomoto, T., B. J. Hoskins, and Y. Matsuda, 2003: The formation mechanism of the Bonin high in August. Quart. J. Roy. Meteor. Soc., 129, 157–178, https://doi.org/10.1256/qj.01.211.
Feng, L., and T. Zhou, 2012: Water vapor transport for summer precipitation over the Tibetan Plateau: Multidata set analysis. J. Geophys. Res., 117, D20114, https://doi.org/10.1029/2011JD017012.
Francis, J. A., and S. J. Vavrus, 2012: Evidence linking Arctic amplification to extreme weather in mid-latitudes. Geophys. Res. Lett., 39, L06801, https://doi.org/10.1029/2012GL051000.
Germe, A., M.-N. Houssais, C. Herbaut, and C. Cassou, 2011: Greenland Sea sea ice variability over 1979–2007 and its link to the surface atmosphere. J. Geophys. Res., 116, C10034, https://doi.org/10.1029/2011JC006960.
Gu, S., Y. Zhang, Q. Wu, and X.-Q. Yang, 2018: The linkage between Arctic sea ice and midlatitude weather: In the perspective of energy. J. Geophys. Res. Atmos., 123, 11 536–11 550, https://doi.org/10.1029/2018JD028743.
Guo, D., Y. Gao, I. Bethke, D. Gong, O. M. Johannessen, and H. Wang, 2013: Mechanism on how the spring Arctic sea ice impacts the East Asian summer monsoon. Theor. Appl. Climatol., 115, 107–119, https://doi.org/10.1007/s00704-013-0872-6.
He, S., Y. Gao, T. Furevik, H. Wang, and F. Li, 2017: Teleconnection between sea ice in the Barents Sea in June and the Silk Road, Pacific-Japan and East Asian rainfall patterns in August. Adv. Atmos. Sci., 35, 52–64, https://doi.org/10.1007/s00376-017-7029-y.
He, Y., and Coauthors, 2021: The mechanism of increasing summer water vapor over the Tibetan Plateau. J. Geophys. Res. Atmos., 126, e2020JD034166, https://doi.org/10.1029/2020JD034166.
Hersbach, H., and Coauthors, 2020: The ERA5 global reanalysis. Quart. J. Roy. Meteor. Soc., 146, 1999–2049, https://doi.org/10.1002/qj.3803.
Hong, X., R. Lu, S. Chen, and S. Li, 2022: The relationship between the North Atlantic Oscillation and the Silk Road pattern in summer. J. Climate, 35, 6691–6702, https://doi.org/10.1175/JCLI-D-21-0833.1.
Hurrell, J. W., and Coauthors, 2013: The Community Earth System Model: A framework for collaborative research. Bull. Amer. Meteor. Soc., 94, 1339–1360, https://doi.org/10.1175/BAMS-D-12-00121.1.
Kitoh, A., and O. Arakawa, 2016: Reduction in the east–west contrast in water budget over the Tibetan Plateau under a future climate. Hydrol. Res. Lett., 10, 113–118, https://doi.org/10.3178/hrl.10.113.
Li, H., H. Chen, H. Wang, J. Sun, and J. Ma, 2018: Can Barents Sea ice decline in spring enhance summer hot drought events over northeastern China? J. Climate, 31, 4705–4725, https://doi.org/10.1175/JCLI-D-17-0429.1.
Lindzen, R. S., and B. Farrell, 1980: A simple approximate result for the maximum growth rate of baroclinic instabilities. J. Atmos. Sci., 37, 1648–1654, https://doi.org/10.1175/1520-0469(1980)037<1648:ASARFT>2.0.CO;2.
Liu, Y., H. Chen, and X. Hu, 2021a: The unstable relationship between the precipitation dipole pattern in the Tibetan Plateau and summer NAO. Geophys. Res. Lett., 48, e2020GL091941, https://doi.org/10.1029/2020GL091941.
Liu, Y., H. Chen, H. Li, G. Zhang, and H. Wang, 2021b: What induces the interdecadal shift of the dipole patterns of summer precipitation trends over the Tibetan Plateau? Int. J. Climatol., 41, 5159–5177, https://doi.org/10.1002/joc.7122.
Ma, Q., J. Zhang, A. T. Game, Y. Chang, and S. Li, 2020: Spatiotemporal variability of summer precipitation and precipitation extremes and associated large-scale mechanisms in Central Asia during 1979–2018. J. Hydrol. X, 8, 100061, https://doi.org/10.1016/j.hydroa.2020.100061.
Ma, Y., G. Tang, D. Long, B. Yong, L. Zhong, W. Wan, and Y. Hong, 2016: Similarity and error intercomparison of the GPM and its predecessor-TRMM multisatellite precipitation analysis using the best available hourly gauge network over the Tibetan Plateau. Remote Sens., 8, 569, https://doi.org/10.3390/rs8070569.
Matsumura, S., and K. Yamazaki, 2012: A longer climate memory carried by soil freeze-thaw processes in Siberia. Environ. Res. Lett., 7, 045402, https://doi.org/10.1088/1748-9326/7/4/045402.
Nakamura, T., K. Yamazaki, T. Sato, and J. Ukita, 2019: Memory effects of Eurasian land processes cause enhanced cooling in response to sea ice loss. Nat. Commun., 10, 5111, https://doi.org/10.1038/s41467-019-13124-2.
Peng, S., W. A. Robinson, and S. Li, 2003: Mechanisms for the NAO responses to the North Atlantic SST tripole. J. Climate, 16, 1987–2004, https://doi.org/10.1175/1520-0442(2003)016<1987:MFTNRT>2.0.CO;2.
Qin, M., S. Li, Y. Xue, and Z. Han, 2022: Intraseasonal variability modes of winter surface air temperature over central Asia and their modulation by Greenland Sea ice and central Pacific El Niño-Southern Oscillation. Int. J. Climatol., 42, 8040–8055, https://doi.org/10.1002/joc.7691.
Qiu, J., 2008: China: The third pole. Nature, 454, 393–396, https://doi.org/10.1038/454393a.
Rayner, N. A., D. E. Parker, E. B. Horton, C. K. Folland, L. V. Alexander, D. P. Rowell, E. C. Kent, and A. Kaplan, 2003: Global analyses of sea surface temperature, sea ice, and night marine air temperature since the late nineteenth century. J. Geophys. Res., 108, 4407, https://doi.org/10.1029/2002JD002670.
Rodell, M., and Coauthors, 2004: The Global Land Data Assimilation System. Bull. Amer. Meteor. Soc., 85, 381–394, https://doi.org/10.1175/BAMS-85-3-381.
Screen, J. A., 2017: Simulated atmospheric response to regional and pan-Arctic sea ice loss. J. Climate, 30, 3945–3962, https://doi.org/10.1175/JCLI-D-16-0197.1.
Sun, J., K. Yang, W. Guo, Y. Wang, J. He, and H. Lu, 2020: Why has the inner Tibetan Plateau become wetter since the mid-1990s? J. Climate, 33, 8507–8522, https://doi.org/10.1175/JCLI-D-19-0471.1.
Sutton, R. T., W. A. Norton, and S. P. Jewson, 2000: The North Atlantic oscillation—What role for the ocean? Atmos. Sci. Lett., 1, 89–100, https://doi.org/10.1006/asle.2000.0021.
Takaya, K., and H. Nakamura, 2001: A formulation of a phase-independent wave-activity flux for stationary and migratory quasigeostrophic eddies on a zonally varying basic flow. J. Atmos. Sci., 58, 608–627, https://doi.org/10.1175/1520-0469(2001)058<0608:AFOAPI>2.0.CO;2.
Tian, L., V. Masson-Delmotte, M. Stievenard, T. Yao, and J. Jouzel, 2001: Tibetan Plateau summer monsoon northward extent revealed by measurements of water stable isotopes. J. Geophys. Res., 106, 28 081–28 088, https://doi.org/10.1029/2001JD900186.
Trenberth, K. E., 1991: Climate diagnostics from global analyses: Conservation of mass in ECMWF analyses. J. Climate, 4, 707–722, https://doi.org/10.1175/1520-0442(1991)004<0707:CDFGAC>2.0.CO;2.
Wang, B., Q. Bao, B. Hoskins, G. Wu, and Y. Liu, 2008: Tibetan Plateau warming and precipitation changes in East Asia. Geophys. Res. Lett., 35, L14702, https://doi.org/10.1029/2008GL034330.
Wu, B., R. Zhang, and B. Wang, 2009: On the association between spring Arctic sea ice concentration and Chinese summer rainfall: A further study. Adv. Atmos. Sci., 26, 666–678, https://doi.org/10.1007/s00376-009-9009-3.
Wu, G., H. Zhuo, Z. Wang, and Y. Liu, 2016: Two types of summertime heating over the Asian large-scale orography and excitation of potential-vorticity forcing I. Over Tibetan Plateau. Sci. China Earth Sci., 59, 1996–2008, https://doi.org/10.1007/s11430-016-5328-2.
Wu, Y., H. Zheng, B. Zhang, D. Chen, and L. Lei, 2014: Long-term changes of lake level and water budget in the Nam Co Lake Basin, central Tibetan Plateau. J. Hydrometeor., 15, 1312–1322, https://doi.org/10.1175/JHM-D-13-093.1.
Xu, K., L. Zhong, Y. Ma, M. Zou, and Z. Huang, 2020: A study on the water vapor transport trend and water vapor source of the Tibetan Plateau. Theor. Appl. Climatol., 140, 1031–1042, https://doi.org/10.1007/s00704-020-03142-2.
Yang, K., B. Ye, D. Zhou, B. Wu, T. Foken, J. Qin, and Z. Zhou, 2011: Response of hydrological cycle to recent climate changes in the Tibetan Plateau. Climatic Change, 109, 517–534, https://doi.org/10.1007/s10584-011-0099-4.
Yang, K., H. Wu, J. Qin, C. Lin, W. Tang, and Y. Chen, 2014: Recent climate changes over the Tibetan Plateau and their impacts on energy and water cycle: A review. Global Planet. Change, 112, 79–91, https://doi.org/10.1016/j.gloplacha.2013.12.001.
Yasui, S., and M. Watanabe, 2010: Forcing processes of the summertime circumglobal teleconnection pattern in a dry AGCM. J. Climate, 23, 2093–2114, https://doi.org/10.1175/2009JCLI3323.1.
Yu, J., Q. Li, Y. Ding, J. Zhang, Q. Wu, and X. Shen, 2022: Long-term trend of water vapor over the Tibetan Plateau in boreal summer under global warming. Sci. China Earth Sci., 65, 662–674, https://doi.org/10.1007/s11430-021-9874-0.
Zhang, G., and Coauthors, 2020: Response of Tibetan Plateau lakes to climate change: Trends, patterns, and mechanisms. Earth-Sci. Rev., 208, 103269, https://doi.org/10.1016/j.earscirev.2020.103269.
Zhang, J., R. Hu, Q. Ma, and M. Niu, 2022: The warming of the Arabian Sea induced a northward summer monsoon over the Tibetan Plateau. J. Climate, 35, 7541–7554, https://doi.org/10.1175/JCLI-D-22-0273.1.
Zhang, L., F. Su, D. Yang, Z. Hao, and K. Tong, 2013: Discharge regime and simulation for the upstream of major rivers over Tibetan Plateau. J. Geophys. Res. Atmos., 118, 8500–8518, https://doi.org/10.1002/jgrd.50665.
Zhang, R., C. Sun, J. Zhu, R. Zhang, and W. Li, 2020: Increased European heat waves in recent decades in response to shrinking Arctic sea ice and Eurasian snow cover. npj Climate Atmos. Sci., 3, 7, https://doi.org/10.1038/s41612-020-0110-8.
Zhao, Y., and T. Zhou, 2019: Asian water tower evinced in total column water vapor: A comparison among multiple satellite and reanalysis data sets. Climate Dyn., 54, 231–245, https://doi.org/10.1007/s00382-019-04999-4.
Zhou, C., P. Zhao, and J. Chen, 2019: The interdecadal change of summer water vapor over the Tibetan Plateau and associated mechanisms. J. Climate, 32, 4103–4119, https://doi.org/10.1175/JCLI-D-18-0364.1.
Zhu, Y., Y. Ding, and H. Xu, 2008: Decadal relationship between atmospheric heat source and winter-spring snow cover over the Tibetan Plateau and rainfall in East China. Acta Meteor. Sin., 22, 303–316.
