Quantifying effects of compound dry-hot extremes on vegetation in Xinjiang (China) using a vine-copula conditional probability model
Graphical abstracts
Introduction
Climate change has aggravated the frequency of weather and climate extreme events (e.g., drought and heatwave) around the world (Trenberth et al., 2014; Touma et al., 2015; Le et al., 2020). Increasing extreme drought and high temperature would largely influence the vegetation growth conditions, such as declining soil moisture and increasing evapotranspiration. The weakened vegetation growth conditions would cause irreversible damage to the structures and functions of terrestrial ecosystems, such vegetation photosynthesis (Mohammat et al., 2013) and nutrient cycling (Cremonese et al., 2017; Berdugo et al., 2020). Consequently, vegetation coverage may tend to decline and bare ground would expend. Such influence may be further aggravated by compound dry-hot extreme events that are caused by the conditional dependence between drought and high temperature. Therefore, vegetation vulnerability under extreme dry-hot events, especially in semi-arid and arid regions, has aroused worldwide attentions. Quantifying the vegetation's response to compound dry-hot extreme events is of great significance for further understanding of the vegetation's response mechanism to climate extremes and formulating effective management for natural ecosystems to climate change.
Over the past decades, a number of studies related to impact assessments of drought and/or hot extremes on vegetation growth have been conducted based on deterministic and probabilistic approaches (Bento et al., 2018; Guo et al., 2018; Zhang and Zhang, 2019; Karimi et al., 2020). The impact of drought and hot extremes on vegetation is generally analyzed through integrating indices of drought (e.g., standardized precipitation index, SPI; standardized precipitation evapotranspiration index, SPEI), temperature (standardized temperature index, STI) and vegetation (standardized difference vegetation index, NDVI) (McKee et al., 1993; Vicente-Serrano et al., 2010; Zscheischler et al., 2014; Zellweger et al., 2019). For example, Xu et al. (2018) discussed the response of NDVI to SPEI under different climatic conditions and land cover types, results would enhance understanding the dominant impacts of different vegetation types to drought scales. Corona-Lozada et al. (2019) used the standard major axis regression to analyze the effects of extreme events on terrestrial ecosystems in the French Alps; results indicated that summer water balance played a key role in vegetation response to heatwaves in mountain grasslands. Jha et al. (2019) evaluated the conditional probabilities of vegetation drought impacted by precipitation, temperature and soil moisture in India with the copula method, results disclosed that the cropland vegetation types were most susceptible to drought events. Generally, most of the previous studies focus on the relationship between vegetation and extreme indicators (drought or hot extremes) individually, without jointly considering their interactive effects.
Recently, the copula method was increasingly employed to model dependencies between extreme events (Chen et al., 2019; Hao et al., 2020; Poonia et al., 2021) and to assess the likelihood of vegetation activity corresponding to extreme events, mainly by establishing joint distribution of bivariate variables (Fang et al., 2019a; Das et al., 2020; Jha et al., 2020). However, due to the heterogeneous dependence among different variable pairs, the multi-parameter copulas are often inflexible in the high-dimensional system modeling, which makes the copula unable to flexibly construct the dependent structures of compound events (Bevacqua et al., 2017). Encouragingly, the vine-copula approach, which decomposes the complex dependent structures into bivariate copulas through pair-copula constructions (PCCs), provides greater flexibility in modeling general high-dimensional systems compared to multi-parameter copulas. Therefore, a potential vine copula approach can better fill the knowledge gap in quantifying the vegetation response through high-dimensional models.
Xinjiang province, as an important region along the Belt and Road in China, is experiencing drought and hot events in recent years, which is the main driving for vegetation degradation. Recent research works have paid efforts in exploring the impacts of drought and hot extremes on soil erosion, land degradation and ecosystem damage for Xinjiang province (Zhang et al., 2015; Li et al., 2018; Yao et al., 2019). These studies are devoted to provide useful information for local ecosystem maintenance and improvement. For example, Hua et al. (2017) explored the relationships between the drought occurrences and vegetation activity in the growing seasons using correlation analysis method, results showed that the occurrence of drought events had significant impacts on vegetation activity during the growing season in semi-arid regions. Li et al. (2018) jointly studied the droughts and hot extremes in northwest China by analyzing the spatial trend of SPEI, results showed that uneven humid trend led to the change of the spatial distribution and trend of droughts and hot extremes in northwest China.
Generally, most of the previous studies mainly focused on the qualitative impact analysis of extreme events on vegetation ecosystems individually; however, they have difficulties in quantitative assessing the impacts of compound extremes (e.g., dry-hot extremes) on vegetation vulnerability. Through the proposal of copula method could model the dependencies between two variables (e.g., NDVI and STI), compound events with high-dimension interactions may beyond the functions of conventional copulas. Besides, as an important strategic region, few studies have been found devoting in model the complex relationships among vegetation and compound dry-hot extremes with vine-copula method. In this regard, it is becoming increasingly urgent to quantitatively assess the potential impacts of compound dry-hot extremes on vegetation growth in Xinjiang.
Therefore, the objective of this study is to develop a vine-copula conditional probability (VCCP) model to analyze the impacts of compound dry-hot extremes on vegetation dynamics. Based on the copula conditional probability method (i.e. pair-copula decompositions), a high-dimensional dependency relationship among the vegetation dynamic and compound extremes is constructed using SPEI, STI and NDVI, respectively. The proposed VCCP model can (i) build the dependence structure of SPEI, STI and NDVI by the pair-copula constructions, (ii) acquire probabilistic response of vegetation loss conditioned on the compound dry-hot extremes, and (iii) disclose the impacts of combined extreme events on different vegetation land-use types. The proposed model is applied to Xinjiang province, where the spatial patterns of correlations among SPEI, STI and NDVI are recognized in the summer season (i.e. June, July and August) and, then, joint distributions of the three indices are established based on the PCC theory. Results will help to enhance the understanding of the climatic extreme impacts on vegetation loss, thus supporting effective management for natural ecosystems to climate change.
Section snippets
Study area
Xinjiang province (between 73°40′ – 96°23′ E and 34°25′ – 49°10′ N), located in the innermost center of Eurasia, is the largest provincial administrative division in China, covering an area of about 1,660,000 km2. It has a typical mountain-basin topography, with mountains and basins interwoven, which is metaphorically called “Two basins lying between three mountain ranges” (as shown in Fig. 1). Two basins are the Junggar Basin in the north and the Tarim Basin in the south, and three mountain
Methodology
In this study, a VCCP model for vegetation vulnerability assessment under compound dry-hot extremes is proposed. Fig. 2 shows the framework of the VCCP model.
Spatial correlation patterns among NDVI, SPEI and STI
Figs. 3, 4 and 5 present the spatial patterns of Pearson correlation coefficients and the corresponding p-values among the three indices in June-August in Xinjiang. in June, NDVI, SPEI and STI are significantly related to each other in most vegetation-covered areas. The dependence between NDVI and SPEI almost presents positive pattern with the 58.1%, 77.0% and 75.8% of the whole pixels, and the percentage of paired NDVI-SPEI correlation at the 5% significance level (p < 0.05) is 11.6%, 22.4%
Conclusions
In this study, a vine copula-based conditional probability (VCCP) framework is proposed to quantify the response of vegetation vulnerability to the compound drought and hot extremes in the summer season in Xinjiang from 1983 to 2015. Firstly, through calculating the correlations among NDVI, SPEI and STI, the spatiotemporal response patterns of vegetation conditions to drought and hot events are identified, respectively. Then, the joint distribution of NDVI, SPEI and STI is constructed by the
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
This research is supported by the Strategic Priority Research Program of Chinese Academy of Sciences (XDA20060302), and the Natural Science Foundation of China (51779008). The authors are grateful to the editors and the anonymous reviewers for their insightful comments and suggestions.
References (52)
- et al.
Pair-copula constructions of multiple dependence
Insurance Math. Econ.
(2009) - et al.
A climatological assessment of drought impact on vegetation health index
Agric. For. Meteorol.
(2018) - et al.
Compound hot droughts over China: Identification, risk patterns and variations
Atmos. Res.
(2019) - et al.
Drought offsets the positive effect of summer heat waves on the canopy greenness of mountain grasslands
Agric. For. Meteorol.
(2019) - et al.
Heat wave hinders green wave: The impact of climate extreme on the phenology of a mountain grassland
Agric. For. Meteorol.
(2017) - et al.
Non-stationary and copula-based approach to assess the drought characteristics encompassing climate indices over the Himalayan states in India
J. Hydrol.
(2020) - et al.
Trends in fall phenology across the deciduous forests of the eastern USA
Agric. For. Meteorol.
(2012) - et al.
Bivariate probabilistic quantification of drought impacts on terrestrial vegetation dynamics in mainland china
J. Hydrol.
(2019) - et al.
Probabilistic assessment of remote sensing-based terrestrial vegetation vulnerability to drought stress of the Loess Plateau in China
Remote Sens. Environ.
(2019) - et al.
Space-time characterization of drought events and their impacts on vegetation in Central Asia
J. Hydrol.
(2018)
Impact of dependence changes on the likelihood of hot extremes under drought conditions in the United States
J. Hydrol.
Probabilistic evaluation of vegetation drought likelihood and its implications to resilience across India
Global Planetary Change
Quantitative assessment and driving force analysis of vegetation drought risk to climate change: Methodology and application in Northeast China
Agric. For. Meteorol.
Climate warming-induced drought constrains vegetation productivity by weakening the temporal stability of the plant community in an arid grassland ecosystem
Agric. For. Meteorol.
Temporal-spatial variations and influencing factors of vegetation cover in Xinjiang from 1982 to 2013 based on GIMMS-NDVI 3g
Global Planet. Change
Drought and spring cooling induced recent decrease in vegetation growth in Inner Asia
Agric. For. Meteorol.
Vine copulas with asymmetric tail dependence and applications to financial return data
Comput. Stat. Data Anal.
Hydraulic constraints on plant gas exchange
Agric. For. Meteorol.
A multi-model and multi-index evaluation of drought characteristics in the 21st century
J. Hydrol.
Diverse responses of vegetation growth to meteorological drought across climate zones and land biomes in northern China from 1981 to 2014
Agric. For. Meteorol.
Hydro-climatic changes and their impacts on vegetation in Xinjiang, Central Asia
Sci. Total Environ.
Advances in microclimate ecology arising from remote sensing
Trends Ecol. Evol.
Grassland vegetation phenological variations and responses to climate change in the Xinjiang region, china
Quat. Int.
The responses of natural vegetation dynamics to drought during the growing season across China
J. Hydrol.
Crop Evapotranspiration: Guidelines for Computing Crop Water Requirements. Irrigation and Drainage Paper Number 56
Vines - A new graphical model for dependent random variables
Annal. Stat.
Cited by (31)
A framework to evaluate the impact of a hazard chain and geographical covariates on spatial extreme water levels: A case study in the Pearl River Delta
2024, Science of the Total EnvironmentContrasting impact of extreme soil and atmospheric dryness on the functioning of trees and forests
2024, Science of the Total EnvironmentThe characteristic of compound drought and saltwater intrusion events in the several major river estuaries worldwide
2024, Journal of Environmental ManagementThe negative-positive feedback transition thresholds of meteorological drought in response to agricultural drought and their dynamics
2024, Science of the Total EnvironmentIncreasing compound drought and hot event over the Tibetan Plateau and its effects on soil water
2023, Ecological Indicators