Characteristics of Drought Events and Their Impact on Vegetation Dynamics in the Arid Region of Northwest China

Guixiang Zhou, Junqiang Yao, Jing Chen, Yaning Chen, Chuan Wang, Yinxue Mo, Chenzhi Ma, Yang Yuhui, Moyan Li, Peng ZhengForests 2024 · DOI

TL;DR

This study investigates how drought events and climate factors influence vegetation health in China's arid Northwest. It reveals that while temperature and precipitation are increasing, the region is experiencing more frequent and severe droughts, with soil moisture and temperature emerging as the most critical drivers of vegetation change.

What Problem Does This Paper Address?

The arid region of Northwest China (ARNC) is particularly vulnerable to climate change, experiencing frequent drought events. Understanding how these droughts, alongside other meteorological factors, impact vegetation dynamics is crucial for predicting future ecosystem health and developing effective conservation strategies. Previous research in this region has been limited in its scope, often focusing on single factors or shorter time periods. This paper aims to provide a comprehensive analysis of the spatiotemporal trends of drought, meteorological conditions, and vegetation cover, and to elucidate the complex interplay between them.

Methods

The researchers employed a multi-faceted approach:

  • Drought Event Characterization: They used the Standardized Precipitation Evapotranspiration Index (SPEI), specifically the 6-month scale, to identify drought events. A "run theory" approach was used to define drought events as periods where SPEI dropped below -1 for more than two months. Key characteristics analyzed included drought times (frequency), duration, severity (cumulative SPEI deficit), and intensity (average SPEI during the event).
  • Trend Analysis: The Theil-Sen analysis and Mann-Kendall (MK) test were used to assess long-term trends in meteorological factors (temperature, precipitation), drought stress factors (vapor pressure deficit, soil moisture), and vegetation cover (NDVI).
  • Impact Analysis: Pearson correlation analysis was used to quantify the linear relationships between different factors and vegetation change. To further disentangle direct and indirect effects, a Structural Equation Model (SEM) was employed.

Datasets

The study utilized several publicly available datasets:

  • Meteorological Data: Climate Research Unit (CRU) time-series (TS) data version 4.06 for temperature (TMP) and precipitation (PRE).
  • Drought Index: Standardized Precipitation Evapotranspiration Index (SPEI) data from the SPEIbase v2.10.
  • Drought Stress Factors:
    • Soil Moisture (SM) data from the Global Land Evaporation Amsterdam Model (GLEAM) version 3.6.
  • Vapor Pressure Deficit (VPD) was calculated from TMP and actual vapor pressure data.
  • Vegetation Data: Normalized Difference Vegetation Index (NDVI) from the PKU GIMMS NDVI dataset (version 1.2), which integrates AVHRR and MODIS data.
  • Land Use Data: Derived from the Resource and Environment Data Cloud Platform for China.

All datasets were resampled to a consistent spatial resolution to enable direct comparison.

Key Insights and Ideas

  • Conflicting Trends: While temperature and precipitation are increasing (a "warming and wetting" trend), the SPEI shows a decreasing trend, indicating a paradoxical increase in drought conditions. This highlights the importance of considering evapotranspiration in drought assessment.
  • Dominant Drivers: The study strongly suggests that soil moisture (SM) and temperature (TMP) are the most influential factors driving vegetation change in the ARNC, exerting greater impacts than precipitation (PRE), vapor pressure deficit (VPD), or SPEI.
  • Interconnectedness: The SEM analysis revealed that TMP indirectly influences vegetation via VPD, and PRE influences it via SM, underscoring the complex, interconnected nature of these environmental factors.
  • Regional Vulnerability: Drought events are not uniformly distributed, with the Tarim River Basin, Turpan-Hami Basin, and Hexi Corridor identified as hotspots for drought occurrence.

Key Results

  • Increasing Trends: Annual and growing season temperature (TMP), precipitation (PRE), vapor pressure deficit (VPD), soil moisture (SM), and NDVI all showed increasing trends between 1982 and 2021.
  • Decreasing Trend: The SPEI exhibited a decreasing trend over the same period, indicating an increase in drought severity.
  • Drought Event Characteristics: Drought events were found to vary significantly across the ARNC, with higher frequencies and intensities in specific basins.
  • Impact on Vegetation: Pearson correlation and SEM analyses consistently showed that SM and TMP had the most significant positive impacts on vegetation growth (NDVI), followed by PRE. VPD and SPEI had less direct or even negative impacts in some cases.

Limitations and Future Work

  • Spatial Resolution: The authors acknowledge that resampling datasets to a consistent spatial resolution might have affected data quality and accuracy.
  • Timescale Differences: The study notes that drought characteristics can vary depending on the timescale analyzed, and different timescales might yield different results.
  • Model Complexity: While SEM provided valuable insights, the authors suggest that future research should explore more complex physical models to better describe the intricate interactions between meteorological factors, drought stress, and vegetation change.
  • Coupled Effects: The study highlights the importance of the coupled effects of VPD and SM on vegetation, suggesting this as a key area for future investigation.

Summary updated

Abstract

The arid region of Northwest China (ARNC) is responsive to global climate change, and drought events have occurred frequently in recent decades. However, studies about the effect of meteorological and drought stress on vegetation change in the ARNC are still insufficient. In this study, we analyzed the spatiotemporal trends of meteorological factors (temperature, TMP; precipitation, PRE; standardized precipitation evapotranspiration index, SPEI), drought stress factors (vapor pressure deficit, VPD; soil moisture, SM), and vegetation (normalized difference vegetation index, NDVI) during 1982–2021. We also investigated the characteristics of drought events by the run theory, including drought times, drought duration, drought severity, and drought intensity. The impacts of meteorological and drought stress factors on the vegetation were explored using Pearson correlation analysis and the structural equation model (SEM). We found that the annual and growing season TMP, PRE, VPD, SM, and NDVI showed an increasing trend in the ARNC during 1982–2021. In contrast, SPEI exhibited a decreasing trend in the annual and growing season. In addition, the characteristics of the drought events varied significantly in the ARNC. The drought events primarily occurred in the Tarim River Basin, Turpan-Hami Basin, and the Hexi Corridor. The Pearson correlation analysis and SEM results consistently demonstrated that TMP and SM exerted greater impacts on vegetation growth than PRE, VPD, and SPEI. The factors that determine vegetation change were TMP and PRE. Exploration of meteorological and drought stress factors that influence vegetation change is essential for comprehending the influence of dominant factors on vegetation change.
BibTeX
@inproceedings{zhou2024characteristics,
  author = {Guixiang Zhou and Junqiang Yao and Jing Chen and Yaning Chen and Chuan Wang and Yinxue Mo and Chenzhi Ma and Yang Yuhui and Moyan Li and Peng Zheng},
  title = {{Characteristics of Drought Events and Their Impact on Vegetation Dynamics in the Arid Region of Northwest China}},
  booktitle = {Forests},
  year = {2024},
  doi = {10.3390/f15122187},
}
Figure