Water availability regulates abrupt shifts in global terrestrial vegetation productivity by reducing resilience
Data files
Jan 27, 2026 version files 8.38 MB
-
README.md
1.24 KB
-
Supplementary_Data.zip
8.38 MB
Abstract
Abrupt shifts in vegetation productivity induced by environmental change profoundly affect ecosystem functions and services, such as carbon sequestration, biodiversity maintenance, and climate regulation, yet their long-term global patterns and underlying drivers remain insufficiently understood. In this study, abrupt shifts in vegetation productivity were identified using two independent long-term satellite datasets, and early warning signals were assessed through ecosystem resilience. The main drivers of resilience decline were further examined across land cover, climate, and human-activity dimensions. The results showed that negative abrupt shifts were predominantly concentrated in high northern latitudes and equatorial regions, and that both negative and positive abrupt shifts displayed distinct increasing trends after 2002. A sharp decline in ecosystem resilience was also detected during the four years preceding abrupt shifts. Changes in water availability were identified as the primary factor contributing to the reduced resilience of global ecosystems, as evidenced by the driver trajectory, generalized additive models, and convergent cross-mapping methods. This study provides insight into the pervasive occurrence of abrupt shifts in vegetation productivity and the discernible impact of climate change. When projecting vegetation dynamics under future climate change, it is essential to consider the significant uncertainties associated with abrupt shifts.
https://doi.org/10.5061/dryad.wh70rxwx1
The dataset contains all of the original data, as well as all of the data and code for the graphs that produced the main text.
Description of the data and file structure
The data processing and plotting are done using NCL (NCAR Command Language, https://www.ncl.ucar.edu/) and R (https://www.r-project.org/), with NCL being the primary tool and R serving as a supplementary tool.
The missing value for all data is -32767.0.
Sharing/Access information
Data was derived from the sources: The raw data can be found in our article.
Detailed methods for all figures and for data processing can be found in the Supplementary_Data.zip, and each file contains a README (as README.pdf) corresponding to its content. Please refer to the contents of Supplementary_Data.zip for details.
Code/Software
All code requires NCL (NCAR Command Language, https://www.ncl.ucar.edu/) or R (https://www.r-project.org/) to run.
