Global Soil Moisture–Climate Interactions during the Peak Growing Season

Yao Feng aKey Laboratory of Water Cycle and Related Land Surface Processes, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing, China

Search for other papers by Yao Feng in
Current site
Google Scholar
PubMed
Close
,
Hong Wang aKey Laboratory of Water Cycle and Related Land Surface Processes, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing, China

Search for other papers by Hong Wang in
Current site
Google Scholar
PubMed
Close
https://orcid.org/0000-0001-5921-2946
,
Wenbin Liu aKey Laboratory of Water Cycle and Related Land Surface Processes, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing, China

Search for other papers by Wenbin Liu in
Current site
Google Scholar
PubMed
Close
, and
Fubao Sun aKey Laboratory of Water Cycle and Related Land Surface Processes, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing, China
bXinjiang Institute of Ecology and Geography, Chinese Academy of Sciences, Urumqi, China
cAkesu National Station of Observation and Research for Oasis Agro-ecosystem, Akesu, China
dCollege of Resources and Environment, University of Chinese Academy of Sciences, Beijing, China

Search for other papers by Fubao Sun in
Current site
Google Scholar
PubMed
Close
Free access

Abstract

Soil moisture (SM) during the vegetation growing season largely affects plant transpiration and photosynthesis, and further alters the land energy and water balance through its impact on the energy partition into latent and sensible heat fluxes. To highlight the impact of strong vegetation activity, we investigate global SM–climate interactions over the peak growing season (PGS) during 1982–2015 based on multisource datasets. Results suggest widespread positive SM–precipitation (P), SM–evapotranspiration (ET), and negative SM–temperature (T) interactions with non-negligible negative SM–P, SM–ET, and positive SM–T interactions over PGS. Relative to the influence of individual climate factors on SM, the compounding effect of climate factors strengthens SM–climate interactions. Simultaneously, variations of SM are dominated by precipitation from 50°N toward the south, by evapotranspiration from 50°N toward the north, and by temperature over the Sahara, western and central Asia, and the Tibetan Plateau. Importantly, the higher probability of concurrent SM wetness and climate extremes indicates the instant response of SM wetness to extreme climate. In contrast, the resistance of vegetation partially contributes to a consequent slower response of SM dryness to extreme climate. We highlight the significance of the compounding effects of climate factors in understanding SM–climate interaction in the context of strong vegetation activity, and the response of SM wetness and dryness to climate extremes.

© 2023 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

Corresponding authors: Hong Wang, wanghong@igsnrr.ac.cn; Wenbin Liu, liuwb@igsnrr.ac.cn

Abstract

Soil moisture (SM) during the vegetation growing season largely affects plant transpiration and photosynthesis, and further alters the land energy and water balance through its impact on the energy partition into latent and sensible heat fluxes. To highlight the impact of strong vegetation activity, we investigate global SM–climate interactions over the peak growing season (PGS) during 1982–2015 based on multisource datasets. Results suggest widespread positive SM–precipitation (P), SM–evapotranspiration (ET), and negative SM–temperature (T) interactions with non-negligible negative SM–P, SM–ET, and positive SM–T interactions over PGS. Relative to the influence of individual climate factors on SM, the compounding effect of climate factors strengthens SM–climate interactions. Simultaneously, variations of SM are dominated by precipitation from 50°N toward the south, by evapotranspiration from 50°N toward the north, and by temperature over the Sahara, western and central Asia, and the Tibetan Plateau. Importantly, the higher probability of concurrent SM wetness and climate extremes indicates the instant response of SM wetness to extreme climate. In contrast, the resistance of vegetation partially contributes to a consequent slower response of SM dryness to extreme climate. We highlight the significance of the compounding effects of climate factors in understanding SM–climate interaction in the context of strong vegetation activity, and the response of SM wetness and dryness to climate extremes.

© 2023 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

Corresponding authors: Hong Wang, wanghong@igsnrr.ac.cn; Wenbin Liu, liuwb@igsnrr.ac.cn

1. Introduction

Soil moisture (SM) is the direct water pool of ground surface vegetation; low SM availability can substantially constrain plant transpiration and photosynthesis (Seneviratne et al. 2010), reducing terrestrial carbon uptake and food production (van der Molen et al. 2011; Vicente-Serrano et al. 2013). It is also a key quantity in the interplay between land surface and the atmosphere while inducing or modulating temperature (T), precipitation (P) (Hauser et al. 2017), and evapotranspiration (ET), the central climate system process that links the surface water and energy balance (Jung et al. 2010; Teuling et al. 2013). Better quantification of the global SM–climate interactions, particularly over the vegetation growing season, is of great significance to understand crucial factors driving the land–atmosphere feedback and to further reduce substantial risks under climate change.

More recently, SM–climate interactions have been widely investigated based on modeling and observational studies (Feng and Liu 2015; Ford et al. 2018; Baur et al. 2018; Jung et al. 2010; Leutwyler et al. 2021). Low SM is caused by reduced P and/or enhanced ET (Seneviratne et al. 2012); however, increased rainfall does not necessarily promise less intense drought (Chen et al. 2012). Although most of the globe has a moderate to strong positive SM–P feedback (Sehler et al. 2019), nonnegligible negative feedback occurs in dry and wet regions due to the complex SM–ET correlation (Yang et al. 2018). Another important impact of SM is related to T whenever SM limits latent heat flux through decreased ET, leading to an increase in sensible heat flux and thus an increase in T. Simultaneously, increased T leads to higher evaporative demand and thus to an enhanced ET and a further decrease in SM (Seneviratne et al. 2010). The SM–T and SM–P couplings can hardly be separated from the critical role of ET in that terrestrial ET can affect P (Koster et al. 2004) and the associated latent heat flux controlling surface T (Jung et al. 2010). ET mainly includes plant transpiration and bare soil evaporation. High ET reduces SM rapidly (Yuan et al. 2019). SM–ET is the principal interaction that determines the spatial pattern and variation of the SM–P feedback (Wei and Dirmeyer 2012). While SM–climate interactions receive increasing attention, the complexity of feedback and the lack of large-scale observations for relevant processes are becoming apparent (Seneviratne et al. 2010). Improved remotely sensed datasets facilitate a deep insight into SM–climate interactions and their influence on vegetation (Nicolai-Shaw et al. 2017; Jiao et al. 2021; Pinnington et al. 2018; Jha and Srivastava 2018).

Low SM enhances T under dry conditions, but vegetation prevents further drying by reducing ET (Seneviratne and Ciais 2017). Vegetation cannot increase P directly but has an indirect effect by enhancing the moisture recycling process to produce more P (Wang et al. 2021). To further understand SM–climate interactions in the context of vegetation impact, we propose a multisource data-driven quantification of global SM–climate interactions during the peak growing season. Specifically, we aim to explore the driving factors in the key processes of global SM–climate interactions, and to discuss the response of SM wetness/dryness to climate extremes.

2. Data and method

a. Data

To measure soil dryness and wetness, we collected the remotely sensed daily 0.25° × 0.25° soil moisture (unit: m3 m−3) developed by the framework of the European Space Agency Climate Change Initiative program (CCI-SM v06.1; https://www.esa-soilmoisture-cci.org/). The CCI-SM are combined by active and passive microwave satellite observations, available since 1978. It has been validated by global observations and is widely used in previous studies (Yang et al. 2018; Nicolai-Shaw et al. 2017; Herrera‐Estrada and Diffenbaugh 2020). For cross-validation, we also obtained the monthly 0.5° × 0.5° ERA-Interim SM (unit: m3 m−3) developed based on observations and model simulations (ERA-SM; https://climatedataguide.ucar.edu/climate-data/era-interim) from the National Center for Atmospheric Research (NCAR). In the ERA-Interim land surface analysis, if there is a difference between 2-m observations and model state, ERA-SM is uniformly added to or removed from the root zone (first three layers) (Tuinenburg and de Vries 2017). The CCI-SM and ERA-SM are positively correlated across the globe except over the high-latitude Northern Hemisphere. Further analysis is conducted at grids with two SM products significantly (p < 0.05) positively correlated using CCI-SM (see Fig. S1a in the online supplemental material). After the selection, the average number of monthly SM at each selected grid is 146.24, and grids with over 80 monthly records of CCI-SM account for 96.8% of all selected grids. The number of monthly CCI-SM records at these grids is shown in Fig. S1b.

Monthly 0.5° × 0.5° gridded observationally based continent precipitation and average temperature were collected from the Climatic Research Unit (CRU TS v.4.05; https://data.ceda.ac.uk/badc/cru/data/cru_ts/cru_ts_4.05/data/). CRU climate data, available from 1901 to today, are widely used for historical analysis and validation (Hao et al. 2013; Liu et al. 2021). In addition, we used monthly 0.25° × 0.25° evapotranspiration from the Global Land Evaporation Amsterdam Model (GLEAM_v3.3a; https://www.gleam.eu/). GLEAM-ET is calculated as the function of the evaporative demand and soil moisture stress, spanning the period 1980–2018. We also used the biweekly 1/12° × 1/12° Global Inventory Modeling and Mapping Studies–Normalized Difference Vegetation Index (GIMMS-NDVI) generated by the National Oceanic and Atmospheric Administration (https://iridl.ldeo.columbia.edu/SOURCES/.NASA/.ARC/.ECOCAST/.GIMMS/.NDVI3g/.v1p0/) to define the vegetation growing season. The GIMMS-NDVI3g.v1 dataset is assembled from different sensors and spans the period 1981–2015. To match the temporal resolution of CRU climate data, GIMMS-NDVI, CCI-SM, and GLEAM-ET were bilinearly regridded to the CRU grid (0.5° × 0.5°). The biweekly GIMMS-NDVI and daily CCI-SM were averaged to monthly data. Both are commonly used methods to process different spatial and temporal resolutions of multiple datasets. Further interpretation is conducted in the peak of growing season during the overlapping period (1982–2015) of all datasets (Table S1).

Table 1

Three experiments investigating the relationship between SM and climate extremes. Numbers refer to the percentiles of each variable.

Table 1

b. Method

According to Nicolai-Shaw et al. (2017), we define the month with the highest vegetation activity measured by the maximum NDVI value (Fig. 1a) and the surrounding 4 months (total: 5 months) as the peak growing season (PGS) of interest. The higher peak is chosen in the case of the double growing season (e.g., in northern India or South China). Then, P, T, ET, NDVI, and SM are extracted for PGS. Grids with no SM data are excluded from further SM-related analysis. To study SM–climate interactions, apart from the commonly used Pearson correlation (Sehler et al. 2019; McKinnon et al. 2021), we incorporate partial correlation to investigate the independent influence of climate factors on SM before (BPGS), during (PGS), and after PGS (APGS) (Fig. 1b). Both correlation coefficients (r) are assessed at 95% (p < 0.05) confidence level. We also use the variance decomposition method (https://ww2.mathworks.cn/matlabcentral/fileexchange/30091-variance-decomposition) to quantify the contribution of climate (P, T, ET) and vegetation (NDVI) to SM. This method is based on the covariance allocation principle, which can be used for attribution analysis. According to this method, the effect of four driving factors (P, T, ET, and NDVI) on SM can be partitioned into the covariance between four driving factors and SM and that among four driving factors.

Fig. 1.
Fig. 1.

(a) Month with the strongest vegetation activity and (b) different stages of the peak of the growing season. PGS, BPGS, and APGS refer to the 5-month peak of growing season, the start to the peak month of PGS (before PGS), and the peak to the end month of PGS (after PGS).

Citation: Journal of Climate 36, 4; 10.1175/JCLI-D-22-0161.1

In addition, the dependence structure of climate extremes and soil wetness/dryness defined by percentiles are modeled with bivariate copulas. The copulas can overcome the shortcomings of counting the co-occurrence rate of extreme conditions with few samples, which have been widely used to assess the relationship between dependent variables (Zhou et al. 2019). Here, we select the best copula function for SM&P, SM&T, and SM&ET in each grid from five commonly used copulas (Gaussian, Student’s t, Gumbel, Clayton, and Frank copulas) according to the Bayesian information criterion. We then have the joint probability distribution of SM and climate factors (collectively referred to as Var) described by a bivariate copula C as FSM,Var(x, y) = P(SM ≤ x, Var ≤ y) = C[FSM(x), FVar(y)] = C(u, υ), where FSM(x) and FVar(y) are transformed into two uniformly distributed random variables u and υ between 0 and 1, and the marginal distribution functions as FSM(x) = P(SM ≤ x) and FVar(y) = P(Var ≤ y). The conditional probability, computed by the joint and marginal probability as P(SM ≤ x | Var ≤ y ) = P(SM ≤ x, Var ≤ y)/P(Var ≤ y), represents the probability of SM dryness/wetness under given climate conditions. Three experiments investigating the relationship between SM wetness/dryness and climate extremes are conducted as in Table 1.

3. Results

a. Global SM–climate interactions during PGS

Global SM–climate interactions remain similar at different stages of PGS (Fig. 2). Both Pearson and partial r reveal general positive SM–P and SM–ET interactions but a negative SM–T interaction (Fig. 2; see also Fig. S2), which are trivial results acknowledged by previous studies (McKinnon et al. 2021; Seneviratne et al. 2010). Simultaneously, nonnegligible negative SM–P, SM–ET, and positive SM–T interactions are also revealed by Pearson r (Figs. 2a–c). Compared to the well-known positive SM–P, SM–ET, and negative SM–T interactions, we highlight the opposite patterns. Negative SM–P interaction over extremely dry Sahara may be caused by the direct infiltration of precipitation through the sands bypassing the surface zone or the direct evaporation of rain before reaching the surface (Pal et al. 2000). Negative SM–ET interaction is likely when increased ET rapidly reduces SM without persistent precipitation supply (Yuan et al. 2019). SM is supplied with water from the melted snow and ice by elevated temperature in the high latitudes of the Northern Hemisphere, whereas it is influenced by the concurring high precipitation and temperature in tropical monsoon (i.e., South Asia) and tropical savanna (i.e., central South America, Africa, and northern Australia) climate, forming positive SM–T interactions. Without considering the interaction of multiple climate factors, partial r indicates relatively weak but generally positive SM–P and SM–ET and negative SM–T associations (Figs. S2a–c). The differences between Pearson and partial r highlight that the combined effect of climate factors has strengthened SM–climate interactions, which deserves further attention in the context of increasing compound climate extremes (Zscheischler et al. 2020).

Fig. 2.
Fig. 2.

(a)–(c) Pearson correlation r between precipitation (P), temperature (T), evapotranspiration (ET), and soil moisture (SM) and [a(i)–c(ii)] the point density (estimated by the Gaussian kernel density) of climate factors and SM–climate interactions during PGS over 1982–2015. Only those significant at 95% level (p < 0.05) are shown.

Citation: Journal of Climate 36, 4; 10.1175/JCLI-D-22-0161.1

We further investigate the relationship between climatological average P, T, ET, SM, and SM–climate interactions over PGS [Figs. 2a(i)–c(ii); see also Figs. S2a(i)–c(ii)]. Stronger positive SM–P associations occur when P ranges from 0 to 100 mm or SM between 0.2 and 0.3 m3 m−3. Strong negative SM–T interactions are likely when T is between 10° and 20°C and ET within 60–90 mm contributes to stronger SM–ET interaction. We also observe consistent declining areas taken by increasing positive SM–climate interactions. SM–climate interactions by partial and Pearson r suggest that the combined effect of climate factors is particularly influential on negative SM–P and SM–ET and positive SM–T relationships. Different processes dominating SM–climate interactions during PGS require more effort.

Spatially, the strongest positive SM–P and SM–ET interactions occur in central Africa, northern Australia, and South Asia. Over these regions, sufficient precipitation supplies SM and further enhances ET (Figs. 3a,i). Higher ET leading to more precipitation is the most uncertain part of the SM–P feedback given the complex processes involved (Seneviratne et al. 2010). Instead, positive SM–P and negative SM–ET interactions over central Europe, southern Australia, and eastern and central North America may be partly influenced by the strong negative SM–T interaction over these regions. Despite the SM supply from precipitation, higher temperature accelerates the ET process, thus reducing SM rapidly, forming negative SM–T and SM–ET associations (Figs. 3b,ii). Decreased ET further increases sensible heat flux and thus enhances temperature. Notably, positive SM–climate correlations concentrate in the winter-dominant high latitudes of the Northern Hemisphere since snow and ice melted by elevated T supply SM and further increase ET, providing water vapor for precipitation formation (Figs. 3c,iii). The key factors driving those processes in land–atmosphere feedback require further discussion.

Fig. 3.
Fig. 3.

Schematic process contributing to SM–climate interactions. Red (blue) arrows suggest positive (negative) interactions and gray ones are processes not reflected by our data.

Citation: Journal of Climate 36, 4; 10.1175/JCLI-D-22-0161.1

Contributions of climate and vegetation to SM variations are consistent at different stages of PGS. Only results of PGS are shown (Fig. 4). The contribution of P to SM exceeds 60% over regions from 50°N toward the south except for the Sahara, west and central Asia, and the Tibetan Plateau, over which T contributes the most (Figs. 4a,b). SM over the regions from 50°N toward the north is largely influenced by ET (Fig. 4c). Compared to climate factors, vegetation has a negligible influence on SM throughout PGS (Fig. 4d). Therefore, we have three regions dominated (contribution above 60%) by P, T, and ET (Fig. 4e). We also discuss the substantial compounding effect of climate factors on SM based on the contribution (two factors accounting for over 30%, respectively). The widespread compounding effect comes from P and ET over P- and ET- dominated regions, followed by that of P and T over T-dominated regions (Fig. 4f). This is understandable in that P is the supply of SM and SM is mainly consumed by plant transpiration and bare soil evaporation (ET). More importantly, variations of ET are largely influenced by P across the globe except for the extremely dry Sahara and extremely cold Canada and northern Russia dominated by T (Fig. S3).

Fig. 4.
Fig. 4.

Contribution by (a) precipitation (P), (b) temperature (T), (c) evapotranspiration (ET), and (d) vegetation (NDVI) to soil moisture (SM), as well as (e) the dominating (contribution over 60%) and (f) the influential (contribution over 30%) factors of soil moisture during PGS over 1982–2015.

Citation: Journal of Climate 36, 4; 10.1175/JCLI-D-22-0161.1

b. Probability of extreme SM–climate interactions

The probability of extreme SM–climate interactions is separately analyzed over P, T and ET-dominated regions. Over P-dominated regions, SM wetness and high P occur more frequently than SM dryness and low P (Figs. 5a,b). The concurrence of ET and SM shows a similar pattern in ET-dominated regions (Figs. 5e,f). The joint probability of SM wetness and low T is higher than that of SM dryness and high T (Figs. 5c,d). This is consistent with the conclusion that lower temperature and higher precipitation may promote SM conservation during the growing season (Xie et al. 2020). Overall, the higher probability of SM wetness concurring with climate extremes shows an instant response of SM wetness to higher precipitation and lower temperature. Instead, the relatively lower probability of SM dryness concurring with climate extremes is partly attributed to the resistance of vegetation to lower precipitation and higher temperature before SM gets dry. In addition, the single peak in the probability distribution function of extreme SM–P and SM–T interactions is consistent with the overwhelming positive SM–P and negative SM–T association under extreme climate conditions. Instead, two peaks in the probability distribution function of SM dryness and low ET are attributed to both positive and negative SM–ET interactions. The higher probability of positive SM–ET interaction is caused by the widespread dominating influence of P on ET whereas only limited ET-dominated regions are largely controlled by T (Fig. S3), which triggers negative SM–ET interaction over PGS.

Fig. 5.
Fig. 5.

Joint probability of soil (a)–(c) wetness (SM > SM90) and (d)–(f) dryness (SM < SM10) and climate extremes over P-, T-, and ET-dominated regions over PGS. Joint probability indicates the concurring probability of soil wetness/dryness and climate extremes. Only significant (p < 0.05) r is considered. The violin plots suggest the probability distribution functions of joint probability.

Citation: Journal of Climate 36, 4; 10.1175/JCLI-D-22-0161.1

The marginal probability demonstrates the highest probability of SM wetness/dryness over P-dominated regions followed by a higher probability of SM wetness over T-dominated regions and that of SM dryness over ET-dominated regions (Figs. 6a,b). Therefore, SM wetness/dryness is largely influenced by precipitation at both high and low ends in that the impact of precipitation on SM is direct (Wei et al. 2008), followed by the influence of SM wetness by lower T and SM dryness by higher ET. The larger range of SM dryness probability over T-dominated regions can be caused by different drying mechanisms in T-dominated regions. For example, the dryness over the Sahara Desert is influenced by northeast trades and the downdraft brought by subtropical high pressure. Dryness over the desert in central Asia is largely influenced by the continental air mass and central Asia is far from oceans and surrounded by mountains and plateaus, which block the transmission of water vapor. At the given climate extremes, the conditional probability suggests a higher probability of SM wetness than dryness (Figs. 6c,d), consistent with the higher joint probability of SM wetness and climate extremes. These findings are of great importance for projecting future SM conditions under climate change over different regions.

Fig. 6.
Fig. 6.

(a),(b) Marginal and (c), (d) conditional probability of SM wetness (SM > SM90) and dryness (SM < SM10) over P-, T-, and ET-dominated regions. Marginal probability indicates the probability of SM wetness and dryness. Conditional probability presents the probability of SM wetness and dryness under given climate extremes. Green triangles stand for the average values.

Citation: Journal of Climate 36, 4; 10.1175/JCLI-D-22-0161.1

4. Discussion

a. SM–vegetation interaction during PGS

Despite the negligible influence of vegetation on SM compared to climate factors, both Pearson and partial r demonstrate the rapid consumption of SM by vegetation during PGS over dry North America, northern Europe, and the extremely dry Sahara (Figs. 7a,i; see also Figs. S4a,i). These regions are characterized by positive SM–P and negative SM–T interactions. According to Jiao et al. (2021), vegetation growth is constrained by water scarcity over the midlatitude Northern Hemisphere and a water surplus can limit vegetation growth in the high latitudes. The slightly positive SM–P correlation over the ET-dominated high-latitude Northern Hemisphere can hardly provide a water surplus condition to constrain vegetation growth; thus, SM from the melted snow and ice support vegetation growth, contributing to a positive SM–NDVI correlation. On the other hand, vegetation growth over the relatively dry T-dominated midlatitude regions can rapidly reduce SM, leading to a negative SM–NDVI correlation. More importantly, the start of growing season before PGS accelerates SM depletion, leading to negative SM–NDVI interaction over northern Europe and North America dominated by the temperate continental climate with low precipitation all year round, which can be partly relieved by the combined effect of climate factors (Figs. 7b,ii; see also Figs. S4b,ii). Simultaneously, higher SM contributes to strong vegetation activity (Chen et al. 2014), which justifies the widespread positive SM–NDVI interaction before PGS. Yu et al. (2017) concluded that positive anomalies of vegetation greenness favor enhanced ET and P across the Sahel (12°–17°N, 20°W–40°E), indicative of amplified moisture recycling. Instead, the positive SM–NDVI interaction after PGS is largely caused by the rapid decrease of SM and NDVI as insufficient SM constrains the vegetation growth (Fig. 7c,ii and Figs. S4c,iii). The differences between Pearson and partial r (SM, NDVI) highlight the significance of SM–climate interactions in understanding SM–vegetation interaction during PGS.

Fig. 7.
Fig. 7.

(a)–(c) Pearson r and (i)–(iii) partial r between vegetation activity (NDVI) and soil moisture (SM) at different stages of PGS over 1982–2015. Only significant (p < 0.05) r is shown.

Citation: Journal of Climate 36, 4; 10.1175/JCLI-D-22-0161.1

b. Uncertainty

The study is subject to certain limitations. The nonlinear response of SM to vegetation activity and climate change determines that SM–climate interactions may be jointly influenced by complex feedbacks and loops instead of the unidirectional effect. The physical process driving the land–atmosphere interaction can differ with geographically differentiated climate zones and vegetation types. We only highlight major processes and driving factors in SM–climate interactions during the prescribed 5-month PGS. In this case, the memory of past precipitation in current SM and the autocorrelation of precipitation are not considered. According to Ford et al. (2018), the precipitation autocorrelation barely influences the results regarding dry or wet soil preference. While we focus on the direct dependency of SM on climate factors, further processes contributing to SM–climate interactions (rainfall infiltration, interception of water on plants, surface roughness, and extreme climate) matter as well. Tuttle and Salvucci (2017) concluded that the statistical determination of causality between soil moisture and precipitation is highly affected by confounding factors. Although we used two SM products (CCI-SM and ERA-SM) to select grids with enough reliable SM records for further analysis, the application of multisource remotely sensed datasets may also bring uncertainty. To ensure robust results, we experimented with different percentiles to define SM dryness/wetness and climate extremes and found that different thresholds only affected the number of climate extremes but barely influenced the response of SM dryness/wetness to climate extremes (Fig. S5). Through the above approaches, we can obtain reliable SM–climate interactions and spatial contribution of climate factors to SM at selected grids. Specific values suggesting those interactions (e.g., correlation, contribution, or probability) may slightly differ with different data source used; however, the known interactions (e.g., positive/negative correlations, the dominating factors or the higher/lower probability) would not be rejected. On the other hand, over those grids with considerable differences between two SM products (correlations are not significant), large uncertainty exists. Therefore, results over those grids are not shown in this study. Although the study is subject to the uncertainty brought by multiple datasets, it still facilitates a better understanding of global SM–climate interactions during the vegetation growing season and provides scientific references for future projection of global soil conditions under climate change.

5. Conclusions

We conduct a multisource data-driven quantification of global SM–climate interactions over the vegetation growing season during 1982–2015. Widespread positive SM–P and SM–ET and negative SM–T interactions are detected over the PGS. The compounding effect of precipitation, temperature, and evapotranspiration strengthens the association of SM with individual factors. Over 60% variance of SM can be justified by precipitation from 50°N toward the south, by temperature over the Sahara, western and central Asia, and the Tibetan Plateau, and by evapotranspiration over regions from 50°N toward the north. More importantly, extreme SM wetness/dryness is largely influenced by precipitation, followed by SM wetness influenced by lower temperature and SM dryness by higher ET. The higher probability of concurrent SM wetness and climate extremes shows the instant response of SM wetness to climate extremes. The lower probability of concurrent SM dryness and climate extremes can be partially explained by the vegetation resistance to unfavorable climate extremes before SM gets dry.

Acknowledgments.

This study was financially supported by the National Natural Science Foundation of China (42022005 and 42001015), the National Research and Development Program of China (2019YFA0606903), the Program for the “Kezhen-Bingwei” Youth Talents (2020RC004 and 2021RC002) from the Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, the China Postdoctoral Science Foundation (2020M670432 and 2021T140657), and the Top-Notch Young Talents Program of China (Fubao Sun).

Data availability statement.

The remotely sensed daily soil moisture is available from the European Space Agency Climate Change Initiative program at https://www.esa-soilmoisture-cci.org/ and the monthly 0.5° × 0.5° reanalysis soil moisture is from ERA-Interim at https://climatedataguide.ucar.edu/climate-data/era-interim. Monthly 0.25° × 0.25° evapotranspiration is from the Global Land Evaporation Amsterdam Model at https://www.gleam.eu/. Monthly 0.5° × 0.5° gridded observationally based continent precipitation and average temperature are available from the Climatic Research Unit at https://data.ceda.ac.uk/badc/cru/data/cru_ts/cru_ts_4.05/data/. The biweekly 1/12° × 1/12° Global Inventory Modeling and Mapping Studies–Normalized Difference Vegetation Index is available at National Oceanic and Atmospheric Administration at https://iridl.ldeo.columbia.edu/SOURCES/.NASA/.ARC/.ECOCAST/.GIMMS/.NDVI3g/.v1p0/.

REFERENCES

  • Baur, F., C. Keil, and G. C. Craig, 2018: Soil moisture–precipitation coupling over central Europe: Interactions between surface anomalies at different scales and the dynamical implication. Quart. J. Roy. Meteor. Soc., 144, 28632875, https://doi.org/10.1002/qj.3415.

    • Search Google Scholar
    • Export Citation
  • Chen, G., and Coauthors, 2012: Drought in the southern United States over the 20th century: Variability and its impacts on terrestrial ecosystem productivity and carbon storage. Climatic Change, 114, 379397, https://doi.org/10.1007/s10584-012-0410-z.

    • Search Google Scholar
    • Export Citation
  • Chen, T., R. A. M. de Jeu, Y. Y. Liu, G. R. van der Werf, A. J. Dolman, 2014: Using satellite based soil moisture to quantify the water driven variability in NDVI: A case study over mainland Australia. Remote Sens. Environ., 140, 330338, https://doi.org/10.1016/j.rse.2013.08.022.

    • Search Google Scholar
    • Export Citation
  • Feng, H., and Y. Liu, 2015: Combined effects of precipitation and air temperature on soil moisture in different land covers in a humid basin. J. Hydrol., 531, 11291140, https://doi.org/10.1016/j.jhydrol.2015.11.016.

    • Search Google Scholar
    • Export Citation
  • Ford, T. W., S. M. Quiring, B. Thakur, R. Jogineedi, A. Houston, S. Yuan, A. Kalra, and N. Lock, 2018: Evaluating soil moisture–precipitation interactions using remote sensing: A sensitivity analysis. J. Hydrometeor., 19, 12371253, https://doi.org/10.1175/JHM-D-17-0243.1.

    • Search Google Scholar
    • Export Citation
  • Hao, Z., A. AghaKouchak, and T. J. Phillips, 2013: Changes in concurrent monthly precipitation and temperature extremes. Environ. Res. Lett., 8, 034014, https://doi.org/10.1088/1748-9326/8/3/034014.

    • Search Google Scholar
    • Export Citation
  • Hauser, M., R. Orth, and S. I. Seneviratne, 2017: Investigating soil moisture–climate interactions with prescribed soil moisture experiments: An assessment with the Community Earth System Model (version 1.2). Geosci. Model Dev., 10, 16651677, https://doi.org/10.5194/gmd-10-1665-2017.

    • Search Google Scholar
    • Export Citation
  • Herrera‐Estrada, J. E., and N. S. Diffenbaugh, 2020: Landfalling droughts: Global tracking of moisture deficits from the oceans onto land. Water Resour. Res., 56, e2019WR026877, https://doi.org/10.1029/2019WR026877.

    • Search Google Scholar
    • Export Citation
  • Jha, S., and R. Srivastava, 2018: Impact of drought on vegetation carbon storage in arid and semi-arid regions. Remote Sens. Appl. Soc. Environ., 11, 2229, https://doi.org/10.1016/j.rsase.2018.04.013.

    • Search Google Scholar
    • Export Citation
  • Jiao, W., L. Wang, W. K. Smith, Q. Chang, H. Wang, and P. D’Odorico, 2021: Observed increasing water constraint on vegetation growth over the last three decades. Nat. Commun., 12, 3777, https://doi.org/10.1038/s41467-021-24016-9.

    • Search Google Scholar
    • Export Citation
  • Jung, M., and Coauthors, 2010: Recent decline in the global land evapotranspiration trend due to limited moisture supply. Nature, 467, 951954, https://doi.org/10.1038/nature09396.

    • Search Google Scholar
    • Export Citation
  • Koster, R. D., and Coauthors, 2004: Regions of strong coupling between soil moisture and precipitation. Science, 305, 11381140, https://doi.org/10.1126/science.1100217.

    • Search Google Scholar
    • Export Citation
  • Leutwyler, D., A. Imamovic, and C. Schär, 2021: The continental-scale soil moisture–precipitation feedback in Europe with parameterized and explicit convection. J. Climate, 34, 53035320, https://doi.org/10.1175/JCLI-D-20-0415.1.

    • Search Google Scholar
    • Export Citation
  • Liu, W., F. Sun, Y. Feng, C. Li, J. Chen, Y.-F. Sang, and Q. Zhang, 2021: Increasing population exposure to global warm-season concurrent dry and hot extremes under different warming levels. Environ. Res. Lett., 16, 094002, https://doi.org/10.1088/1748-9326/ac188f.

    • Search Google Scholar
    • Export Citation
  • McKinnon, K. A., A. Poppick, and I. R. Simpson, 2021: Hot extremes have become drier in the United States Southwest. Nat. Climate Change, 11, 598604, https://doi.org/10.1038/s41558-021-01076-9.

    • Search Google Scholar
    • Export Citation
  • Nicolai-Shaw, N., J. Zscheischler, M. Hirschi, L. Gudmundsson, and S. I. Seneviratne, 2017: A drought event composite analysis using satellite remote-sensing based soil moisture. Remote Sens. Environ., 203, 216225, https://doi.org/10.1016/j.rse.2017.06.014.

    • Search Google Scholar
    • Export Citation
  • Pal, J. S., E. E. Small, and E. A. B. Eltahir, 2000: Simulation of regional-scale water and energy budgets: Representation of subgrid cloud and precipitation processes within RegCM. J. Geophys. Res., 105, 29 57929 594, https://doi.org/10.1029/2000JD900415.

    • Search Google Scholar
    • Export Citation
  • Pinnington, E., T. Quaife, and E. Black, 2018: Impact of remotely sensed soil moisture and precipitation on soil moisture prediction in a data assimilation system with the JULES land surface model. Hydrol. Earth Syst. Sci., 22, 25752588, https://doi.org/10.5194/hess-22-2575-2018.

    • Search Google Scholar
    • Export Citation
  • Sehler, R., J. Li, J. Reager, and H. Ye, 2019: Investigating relationship between soil moisture and precipitation globally using remote sensing observations. J. Contemp. Water Res. Educ., 168, 106118, https://doi.org/10.1111/j.1936-704X.2019.03324.x.

    • Search Google Scholar
    • Export Citation
  • Seneviratne, S. I., and P. Ciais, 2017: Trends in ecosystem recovery from drought. Nature, 548, 164165, https://doi.org/10.1038/548164a.

    • Search Google Scholar
    • Export Citation
  • Seneviratne, S. I., T. Corti, E. L. Davin, M. Hirschi, E. B. Jaeger, I. Lehner, B. Orlowsky, and A. J. Teuling, 2010: Investigating soil moisture–climate interactions in a changing climate: A review. Earth-Sci. Rev., 99, 125161, https://doi.org/10.1016/j.earscirev.2010.02.004.

    • Search Google Scholar
    • Export Citation
  • Seneviratne, S. I., and Coauthors, 2012: Changes in climate extremes and their impacts on the natural physical environment. Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation, C. B. Field et al., Eds., Cambridge University Press, 109–230.

  • Teuling, A. J., and Coauthors, 2013: Evapotranspiration amplifies European summer drought. Geophys. Res. Lett., 40, 20712075, https://doi.org/10.1002/grl.50495.

    • Search Google Scholar
    • Export Citation
  • Tuinenburg, O. A., and J. P. R. de Vries, 2017: Irrigation patterns resemble ERA-interim reanalysis soil moisture additions. Geophys. Res. Lett., 44, 10 34110 348, https://doi.org/10.1002/2017GL074884.

    • Search Google Scholar
    • Export Citation
  • Tuttle, E., and G. D. Salvucci, 2017: Confounding factors in determining causal soil moisture–precipitation feedback. Water Resour. Res., 53, 55315544, https://doi.org/10.1002/2016WR019869.

    • Search Google Scholar
    • Export Citation
  • van der Molen, M. K., and Coauthors, 2011: Drought and ecosystem carbon cycling. Agric. For. Meteor., 151, 765773, https://doi.org/10.1016/j.agrformet.2011.01.018.

    • Search Google Scholar
    • Export Citation
  • Vicente-Serrano, S. M., and Coauthors, 2013: Response of vegetation to drought time-scales across global land biomes. Proc. Natl. Acad. Sci. USA, 110, 5257, https://doi.org/10.1073/pnas.1207068110.

    • Search Google Scholar
    • Export Citation
  • Wang, X., and Coauthors, 2021: Vegetation restoration projects intensify intraregional water recycling processes in the agro-pastoral ecotone of northern China. J. Hydrometeor., 22, 13854103, https://doi.org/10.1175/JHM-D-20-0125.1.

    • Search Google Scholar
    • Export Citation
  • Wei, J., and P. A. Dirmeyer, 2012: Dissecting soil moisture–precipitation coupling. Geophys. Res. Lett., 39, L19711, https://doi.org/10.1029/2012GL053038.

    • Search Google Scholar
    • Export Citation
  • Wei, J., P. A. Dirmeyer, and Z. Guo, 2008: Sensitivities of soil wetness simulation to uncertainties in precipitation and radiation. Geophys. Res. Lett., 35, L15703, https://doi.org/10.1029/2008GL034494.

    • Search Google Scholar
    • Export Citation
  • Xie, Q., J. Li, and Y. Zhao, 2020: Effects of air temperature and precipitation on soil moisture on the Qinghai-Tibet plateau during the 2015 growing season. Adv. Meteor., 2020, 4918945, https://doi.org/10.1155/2020/4918945.

    • Search Google Scholar
    • Export Citation
  • Yang, L., G. Sun, L. Zhi, and J. Zhao, 2018: Negative soil moisture–precipitation feedback in dry and wet regions. Sci. Rep., 8, 4026, https://doi.org/10.1038/s41598-018-22394-7.

    • Search Google Scholar
    • Export Citation
  • Yu, Y., M. Notaro, F. Wang, J. Mao, X. Shi, and Y. Wei, 2017: Observed positive vegetation–rainfall feedbacks in the Sahel dominated by a moisture recycling mechanism. Nat. Commun., 8, 1873, https://doi.org/10.1038/s41467-017-02021-1.

    • Search Google Scholar
    • Export Citation
  • Yuan, X., L. Wang, P. Wu, P. Ji, J. Sheffield, and M. Zhang, 2019: Anthropogenic shift towards higher risk of flash drought over China. Nat. Commun., 10, 4661, https://doi.org/10.1038/s41467-019-12692-7.

    • Search Google Scholar
    • Export Citation
  • Zhou, S., and Coauthors, 2019: Land–atmosphere feedbacks exacerbate concurrent soil drought and atmospheric aridity. Proc. Natl. Acad. Sci. USA, 116, 18 84818 853, https://doi.org/10.1073/pnas.1904955116.

    • Search Google Scholar
    • Export Citation
  • Zscheischler, J., B. van den Hurk, P. J. Ward, and S. Westra, 2020: Multivariate extremes and compound events. Climate Extremes and Their Implications for Impact and Risk Assessment, J. Sillmann, S. Sippel, and S. Russo, Eds., Elsevier, 59–76.

Supplementary Materials

Save
  • Baur, F., C. Keil, and G. C. Craig, 2018: Soil moisture–precipitation coupling over central Europe: Interactions between surface anomalies at different scales and the dynamical implication. Quart. J. Roy. Meteor. Soc., 144, 28632875, https://doi.org/10.1002/qj.3415.

    • Search Google Scholar
    • Export Citation
  • Chen, G., and Coauthors, 2012: Drought in the southern United States over the 20th century: Variability and its impacts on terrestrial ecosystem productivity and carbon storage. Climatic Change, 114, 379397, https://doi.org/10.1007/s10584-012-0410-z.

    • Search Google Scholar
    • Export Citation
  • Chen, T., R. A. M. de Jeu, Y. Y. Liu, G. R. van der Werf, A. J. Dolman, 2014: Using satellite based soil moisture to quantify the water driven variability in NDVI: A case study over mainland Australia. Remote Sens. Environ., 140, 330338, https://doi.org/10.1016/j.rse.2013.08.022.

    • Search Google Scholar
    • Export Citation
  • Feng, H., and Y. Liu, 2015: Combined effects of precipitation and air temperature on soil moisture in different land covers in a humid basin. J. Hydrol., 531, 11291140, https://doi.org/10.1016/j.jhydrol.2015.11.016.

    • Search Google Scholar
    • Export Citation
  • Ford, T. W., S. M. Quiring, B. Thakur, R. Jogineedi, A. Houston, S. Yuan, A. Kalra, and N. Lock, 2018: Evaluating soil moisture–precipitation interactions using remote sensing: A sensitivity analysis. J. Hydrometeor., 19, 12371253, https://doi.org/10.1175/JHM-D-17-0243.1.

    • Search Google Scholar
    • Export Citation
  • Hao, Z., A. AghaKouchak, and T. J. Phillips, 2013: Changes in concurrent monthly precipitation and temperature extremes. Environ. Res. Lett., 8, 034014, https://doi.org/10.1088/1748-9326/8/3/034014.

    • Search Google Scholar
    • Export Citation
  • Hauser, M., R. Orth, and S. I. Seneviratne, 2017: Investigating soil moisture–climate interactions with prescribed soil moisture experiments: An assessment with the Community Earth System Model (version 1.2). Geosci. Model Dev., 10, 16651677, https://doi.org/10.5194/gmd-10-1665-2017.

    • Search Google Scholar
    • Export Citation
  • Herrera‐Estrada, J. E., and N. S. Diffenbaugh, 2020: Landfalling droughts: Global tracking of moisture deficits from the oceans onto land. Water Resour. Res., 56, e2019WR026877, https://doi.org/10.1029/2019WR026877.

    • Search Google Scholar
    • Export Citation
  • Jha, S., and R. Srivastava, 2018: Impact of drought on vegetation carbon storage in arid and semi-arid regions. Remote Sens. Appl. Soc. Environ., 11, 2229, https://doi.org/10.1016/j.rsase.2018.04.013.

    • Search Google Scholar
    • Export Citation
  • Jiao, W., L. Wang, W. K. Smith, Q. Chang, H. Wang, and P. D’Odorico, 2021: Observed increasing water constraint on vegetation growth over the last three decades. Nat. Commun., 12, 3777, https://doi.org/10.1038/s41467-021-24016-9.

    • Search Google Scholar
    • Export Citation
  • Jung, M., and Coauthors, 2010: Recent decline in the global land evapotranspiration trend due to limited moisture supply. Nature, 467, 951954, https://doi.org/10.1038/nature09396.

    • Search Google Scholar
    • Export Citation
  • Koster, R. D., and Coauthors, 2004: Regions of strong coupling between soil moisture and precipitation. Science, 305, 11381140, https://doi.org/10.1126/science.1100217.

    • Search Google Scholar
    • Export Citation
  • Leutwyler, D., A. Imamovic, and C. Schär, 2021: The continental-scale soil moisture–precipitation feedback in Europe with parameterized and explicit convection. J. Climate, 34, 53035320, https://doi.org/10.1175/JCLI-D-20-0415.1.

    • Search Google Scholar
    • Export Citation
  • Liu, W., F. Sun, Y. Feng, C. Li, J. Chen, Y.-F. Sang, and Q. Zhang, 2021: Increasing population exposure to global warm-season concurrent dry and hot extremes under different warming levels. Environ. Res. Lett., 16, 094002, https://doi.org/10.1088/1748-9326/ac188f.

    • Search Google Scholar
    • Export Citation
  • McKinnon, K. A., A. Poppick, and I. R. Simpson, 2021: Hot extremes have become drier in the United States Southwest. Nat. Climate Change, 11, 598604, https://doi.org/10.1038/s41558-021-01076-9.

    • Search Google Scholar
    • Export Citation
  • Nicolai-Shaw, N., J. Zscheischler, M. Hirschi, L. Gudmundsson, and S. I. Seneviratne, 2017: A drought event composite analysis using satellite remote-sensing based soil moisture. Remote Sens. Environ., 203, 216225, https://doi.org/10.1016/j.rse.2017.06.014.

    • Search Google Scholar
    • Export Citation
  • Pal, J. S., E. E. Small, and E. A. B. Eltahir, 2000: Simulation of regional-scale water and energy budgets: Representation of subgrid cloud and precipitation processes within RegCM. J. Geophys. Res., 105, 29 57929 594, https://doi.org/10.1029/2000JD900415.

    • Search Google Scholar
    • Export Citation
  • Pinnington, E., T. Quaife, and E. Black, 2018: Impact of remotely sensed soil moisture and precipitation on soil moisture prediction in a data assimilation system with the JULES land surface model. Hydrol. Earth Syst. Sci., 22, 25752588, https://doi.org/10.5194/hess-22-2575-2018.

    • Search Google Scholar
    • Export Citation
  • Sehler, R., J. Li, J. Reager, and H. Ye, 2019: Investigating relationship between soil moisture and precipitation globally using remote sensing observations. J. Contemp. Water Res. Educ., 168, 106118, https://doi.org/10.1111/j.1936-704X.2019.03324.x.

    • Search Google Scholar
    • Export Citation
  • Seneviratne, S. I., and P. Ciais, 2017: Trends in ecosystem recovery from drought. Nature, 548, 164165, https://doi.org/10.1038/548164a.

    • Search Google Scholar
    • Export Citation
  • Seneviratne, S. I., T. Corti, E. L. Davin, M. Hirschi, E. B. Jaeger, I. Lehner, B. Orlowsky, and A. J. Teuling, 2010: Investigating soil moisture–climate interactions in a changing climate: A review. Earth-Sci. Rev., 99, 125161, https://doi.org/10.1016/j.earscirev.2010.02.004.

    • Search Google Scholar
    • Export Citation
  • Seneviratne, S. I., and Coauthors, 2012: Changes in climate extremes and their impacts on the natural physical environment. Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation, C. B. Field et al., Eds., Cambridge University Press, 109–230.

  • Teuling, A. J., and Coauthors, 2013: Evapotranspiration amplifies European summer drought. Geophys. Res. Lett., 40, 20712075, https://doi.org/10.1002/grl.50495.

    • Search Google Scholar
    • Export Citation
  • Tuinenburg, O. A., and J. P. R. de Vries, 2017: Irrigation patterns resemble ERA-interim reanalysis soil moisture additions. Geophys. Res. Lett., 44, 10 34110 348, https://doi.org/10.1002/2017GL074884.

    • Search Google Scholar
    • Export Citation
  • Tuttle, E., and G. D. Salvucci, 2017: Confounding factors in determining causal soil moisture–precipitation feedback. Water Resour. Res., 53, 55315544, https://doi.org/10.1002/2016WR019869.

    • Search Google Scholar
    • Export Citation
  • van der Molen, M. K., and Coauthors, 2011: Drought and ecosystem carbon cycling. Agric. For. Meteor., 151, 765773, https://doi.org/10.1016/j.agrformet.2011.01.018.

    • Search Google Scholar
    • Export Citation
  • Vicente-Serrano, S. M., and Coauthors, 2013: Response of vegetation to drought time-scales across global land biomes. Proc. Natl. Acad. Sci. USA, 110, 5257, https://doi.org/10.1073/pnas.1207068110.

    • Search Google Scholar
    • Export Citation
  • Wang, X., and Coauthors, 2021: Vegetation restoration projects intensify intraregional water recycling processes in the agro-pastoral ecotone of northern China. J. Hydrometeor., 22, 13854103, https://doi.org/10.1175/JHM-D-20-0125.1.

    • Search Google Scholar
    • Export Citation
  • Wei, J., and P. A. Dirmeyer, 2012: Dissecting soil moisture–precipitation coupling. Geophys. Res. Lett., 39, L19711, https://doi.org/10.1029/2012GL053038.

    • Search Google Scholar
    • Export Citation
  • Wei, J., P. A. Dirmeyer, and Z. Guo, 2008: Sensitivities of soil wetness simulation to uncertainties in precipitation and radiation. Geophys. Res. Lett., 35, L15703, https://doi.org/10.1029/2008GL034494.

    • Search Google Scholar
    • Export Citation
  • Xie, Q., J. Li, and Y. Zhao, 2020: Effects of air temperature and precipitation on soil moisture on the Qinghai-Tibet plateau during the 2015 growing season. Adv. Meteor., 2020, 4918945, https://doi.org/10.1155/2020/4918945.

    • Search Google Scholar
    • Export Citation
  • Yang, L., G. Sun, L. Zhi, and J. Zhao, 2018: Negative soil moisture–precipitation feedback in dry and wet regions. Sci. Rep., 8, 4026, https://doi.org/10.1038/s41598-018-22394-7.

    • Search Google Scholar
    • Export Citation
  • Yu, Y., M. Notaro, F. Wang, J. Mao, X. Shi, and Y. Wei, 2017: Observed positive vegetation–rainfall feedbacks in the Sahel dominated by a moisture recycling mechanism. Nat. Commun., 8, 1873, https://doi.org/10.1038/s41467-017-02021-1.

    • Search Google Scholar
    • Export Citation
  • Yuan, X., L. Wang, P. Wu, P. Ji, J. Sheffield, and M. Zhang, 2019: Anthropogenic shift towards higher risk of flash drought over China. Nat. Commun., 10, 4661, https://doi.org/10.1038/s41467-019-12692-7.

    • Search Google Scholar
    • Export Citation
  • Zhou, S., and Coauthors, 2019: Land–atmosphere feedbacks exacerbate concurrent soil drought and atmospheric aridity. Proc. Natl. Acad. Sci. USA, 116, 18 84818 853, https://doi.org/10.1073/pnas.1904955116.

    • Search Google Scholar
    • Export Citation
  • Zscheischler, J., B. van den Hurk, P. J. Ward, and S. Westra, 2020: Multivariate extremes and compound events. Climate Extremes and Their Implications for Impact and Risk Assessment, J. Sillmann, S. Sippel, and S. Russo, Eds., Elsevier, 59–76.

  • Fig. 1.

    (a) Month with the strongest vegetation activity and (b) different stages of the peak of the growing season. PGS, BPGS, and APGS refer to the 5-month peak of growing season, the start to the peak month of PGS (before PGS), and the peak to the end month of PGS (after PGS).

  • Fig. 2.

    (a)–(c) Pearson correlation r between precipitation (P), temperature (T), evapotranspiration (ET), and soil moisture (SM) and [a(i)–c(ii)] the point density (estimated by the Gaussian kernel density) of climate factors and SM–climate interactions during PGS over 1982–2015. Only those significant at 95% level (p < 0.05) are shown.

  • Fig. 3.

    Schematic process contributing to SM–climate interactions. Red (blue) arrows suggest positive (negative) interactions and gray ones are processes not reflected by our data.

  • Fig. 4.

    Contribution by (a) precipitation (P), (b) temperature (T), (c) evapotranspiration (ET), and (d) vegetation (NDVI) to soil moisture (SM), as well as (e) the dominating (contribution over 60%) and (f) the influential (contribution over 30%) factors of soil moisture during PGS over 1982–2015.

  • Fig. 5.

    Joint probability of soil (a)–(c) wetness (SM > SM90) and (d)–(f) dryness (SM < SM10) and climate extremes over P-, T-, and ET-dominated regions over PGS. Joint probability indicates the concurring probability of soil wetness/dryness and climate extremes. Only significant (p < 0.05) r is considered. The violin plots suggest the probability distribution functions of joint probability.

  • Fig. 6.

    (a),(b) Marginal and (c), (d) conditional probability of SM wetness (SM > SM90) and dryness (SM < SM10) over P-, T-, and ET-dominated regions. Marginal probability indicates the probability of SM wetness and dryness. Conditional probability presents the probability of SM wetness and dryness under given climate extremes. Green triangles stand for the average values.

  • Fig. 7.

    (a)–(c) Pearson r and (i)–(iii) partial r between vegetation activity (NDVI) and soil moisture (SM) at different stages of PGS over 1982–2015. Only significant (p < 0.05) r is shown.

All Time Past Year Past 30 Days
Abstract Views 517 293 11
Full Text Views 241 140 14
PDF Downloads 310 164 10