ENSO amplifies global vegetation resilience variability in a changing climate

Abstract

A thorough understanding of vegetation resilience to climate variability is critical for sustaining ecosystem functions and terrestrial carbon sinks. Although the El Niño-Southern Oscillation (ENSO) is a key driver of global extreme weather events and vegetation dynamics, its impacts on vegetation resilience remain unclear. Here we estimate global present-day (1981-2018) and future (2015-2100) vegetation resilience using a lag-1 autocorrelation analysis of global leaf area index (LAI) time series and investigate its teleconnection to ENSO. Our findings reveal that ENSO significantly affects vegetation resilience across 53% of the global vegetated area. Within these regions, 15% are linked primarily to large-scale atmospheric synchrony with ENSO, 51% are mainly shaped by ENSO-driven local climate anomalies, and the remaining 34% are influenced by both processes. Future projections suggest that the area impacted via ENSO-driven climate anomalies may expand by 7-10%, with Eastern Siberia and northern North America newly affected. Our study provides a coherent global assessment of vegetation resilience sensitivity to ENSO, identifies teleconnected hotspots and potential influential pathways, and informs targeted restoration and climate-adaptive ecosystem governance under climate change.

Fig. 1. Global patterns of historical vegetation resilience changes.

a lag-1 autocorrelation (AC) trend and b variance trend derived from Global Land Surface Satellite Leaf Area Index (GLASS LAI) data from 1981 to 2018. Trends are quantified using the Kendall’s tau statistic. Red regions indicate significant increases in AC or variance, which are indicative of declining vegetation resilience. c Trend agreement between AC and variance. d LAI time series trend (1981–2018). e Proportion of autocorrelation increase (ΔAC > 0) or decrease (ΔAC < 0) in greening (ΔLAI > 0) and browning (ΔLAI < 0) among different vegetation types. Vegetation types: ENF Evergreen Needleleaf Forests, EBF Evergreen Broadleaf Forests, DNF Deciduous Needleleaf Forests, DBF Deciduous Broadleaf Forests, MF Mixed Forests, CS Closed Shrublands, OS Open Shrublands, WS Woody Savannas. Basemap data© Esri; Garmin; GEODIS; GMI; CIA World Factbook.

Fig. 2. Influence of explanatory variables on global vegetation resilience from 1981 to 2018.

a Variables’ importance scores from the Random Forest (RF) model. Partial dependence plots (PDPs) show the influence of key predictors from each variable group on lag-1 autocorrelation (AC): b climatic variables, including El Niño–Southern Oscillation (ENSO), precipitation, temperature, and radiation; c Lag effects of ENSO at 1-, 2-, and 3-year intervals; d Climate-mediated teleconnection effects, calculated as the product of ENSO and in-situ climate variables. Positive and negative extremes represent high anomaly values associated with El Niño and La Niña events, respectively, while values near zero indicate low anomalies; e climatic seasonality, represented by intra-annual coefficient of variation (CV) of precipitation, temperature, and radiation (e.g., temperature_seas indicates temperature seasonality); and f topographic factors, including elevation, slope, and Height Above Nearest Drainage (HAND); lower HAND values denote poorer drainage potential. Basemap data ©Esri; Garmin; GEODIS; GMI; CIA World Factbook.

Fig. 3. ENSO impact on global vegetation resilience.

a Synchronization impact of El Niño–Southern Oscillation (ENSO) on resilience. b Regional differences in lag-1 autocorrelation (AC) trends between affected and unaffected regions during 1981–2018. The dashed line represents the regional average. The inset box in the top right corner magnifies the difference in regional averages. Asterisks indicate statistical significance between groups based on both the Whitney U-test and Student’s t-test: p < 0.05(*); p < 0.01(**); p < 0.001(***). Climate-mediated teleconnection effects on vegetation resilience: c–e Impacts of ENSO-induced anomalies in temperature, solar radiation, and precipitation on vegetation resilience, respectively. f, g Combined impact of these climate anomalies during El Niño and La Niña on vegetation resilience, respectively; h Overall impact of ENSO-induced climate anomalies on vegetation resilience. Plus (+) and minus (−) signs indicate positive and negative impacts, respectively, with numbers indicating impact intensity, calculated as the sum of individual ENSO-climate anomaly impacts. Basemap data © Esri; Garmin; GEODIS; GMI; CIA World Factbook.

Fig. 4. Projected vegetation resilience trend and the synchronization and climate-mediated teleconnection impact of El Niño–Southern Oscillation (ENSO) on vegetation resilience under future scenarios (2015–2100).

a Resilience trend under SSP 126, SSP 245, and SSP 370, respectively. ENSO impacts on vegetation resilience under the three scenarios, including b atmospheric synchronization effects and c climate-mediated effects transmitted through ENSO-driven anomalies. Plus (+) and minus (−) signs indicate positive and negative impacts, respectively, with numbers indicating impact intensity, calculated as the sum of individual grid-level impacts. Basemap data © Esri; Garmin; GEODIS; GMI; CIA World Factbook.