TL;DR
Tropical savannas are shifting towards increased woody vegetation, driven by changes in rainfall patterns. This shift is leading to a decreased sensitivity of overall vegetation productivity (LAImax) to annual rainfall, suggesting a fundamental change in how these ecosystems function and respond to climate.
What Problem Does This Paper Address?
Tropical savannas are dynamic ecosystems where trees and grasses constantly compete for water. This competition shapes the landscape. Scientists have observed a phenomenon called "woody encroachment" in many savannas, where trees are becoming more dominant. While factors like increased CO2, grazing, and fire frequency changes have been proposed as drivers, the specific role of altered rainfall patterns, especially extreme events, in this structural shift has been less clear on a global scale. This paper aims to quantify how changes in rainfall climatology are influencing the balance between woody and herbaceous vegetation in tropical savannas worldwide and how this impacts overall ecosystem productivity.
Methods
The researchers employed a multi-faceted approach using satellite data and statistical analysis. They analyzed:
- Vegetation Optical Depth (VOD): This passive microwave satellite data is sensitive to the water content of woody vegetation, making it a good proxy for woody cover.
- Leaf Area Index (LAImax): Derived from optical satellite data, LAImax represents the total green vegetation production, encompassing both woody and herbaceous foliage.
- Rainfall Data: High-resolution daily rainfall estimates were used to analyze trends in annual rainfall, heavy rainfall frequency, and the number of rainy days.
They used statistical methods like Theil-Sen trend analysis to identify significant trends in these variables over time. To understand the drivers of woody cover change, they employed a bootstrapping technique (LMG method) to determine the relative importance of various factors, including rainfall climatology, temperature, solar radiation, and human management. They also calculated the sensitivity of LAImax to annual rainfall (BLAI-Rainfall) using linear regression over a moving window to assess how vegetation productivity responds to rainfall changes.
Datasets
- GIMMS3g LAI v1: A global dataset of Leaf Area Index (LAI) from 1982-2015.
- Vegetation Optical Depth (VOD) data: Derived from passive microwave observations, used as a proxy for woody cover (1992-2012).
- CRU TS 4.04 and CHIRPS: Climate datasets providing monthly rainfall, temperature, and cloud cover (1982-2015) and daily rainfall estimates, respectively.
- Gridded Population of the World (GPWv4): For human population density data.
- Global Fire Emissions Database (GFED4s): For burned area fraction data.
- Harmonized World Soil Database (HWSD v1.2): For soil organic carbon and sand fraction.
- SRTM 90m Digital Elevation Database: For elevation data.
- ESA CCI Land Cover: Used for masking and defining savanna areas.
Key Insights and Ideas
- Woody Vegetation is Outpacing Overall Greenness: The study found that woody vegetation has increased more significantly (5%) than overall green vegetation production (LAImax, 3%) in arid and semi-arid savannas. This suggests a structural shift where trees are becoming more dominant.
- Savannas are Becoming Less Responsive to Rainfall: A widespread decline in the sensitivity of LAImax to annual rainfall was observed across 56% of tropical savannas. This means that even if annual rainfall increases, the overall green vegetation response is diminishing, indicating a change in ecosystem functioning.
- Heavy Rainfall is a Key Driver: Increased heavy rainfall frequency is strongly linked to increased woody cover. This is likely because woody plants are better equipped to utilize infrequent, intense rainfall events compared to grasses.
- Rainfall Climatology is the Dominant Factor: Changes in rainfall climatology (annual rainfall, heavy rainfall frequency, and rainy days) were identified as the primary drivers of woody cover changes, accounting for a significant portion of the observed shifts, especially in arid and semi-arid regions.
Key Results
- Woody Cover Increase: A statistically significant increase in woody vegetation was observed, particularly in arid and semi-arid regions.
- LAImax Sensitivity Decline: Over 56% of tropical savannas showed a reduced sensitivity of LAImax to annual rainfall. This decline was most pronounced in areas experiencing increased woody cover.
- Rainfall Climatology's Influence: Rainfall climatology variables explained a substantial portion of the variance in woody cover changes (47% globally, and up to 78% in arid regions).
- Heavy Rainfall Correlation: Approximately 24% of global savanna regions showed an increase in both woody cover and heavy rainfall frequency.
Limitations and Future Work
- Grazing Data Limitations: The study acknowledges that reliable global data on grazing intensities was not available, preventing its inclusion in the primary analysis of drivers. This is a significant factor in some regions, particularly Southern Africa.
- Human Management Complexity: While human management factors were considered, their complex interactions and the availability of consistent global data present challenges for precise quantification.
- CO2 Fertilization: The paper notes that while CO2 fertilization can contribute to vegetation productivity, its global impact on woody encroachment is difficult to quantify and observational evidence is not yet conclusive. The study emphasizes that altered rainfall climatology appears to be a more dominant global driver.
Future research could focus on integrating more detailed grazing data and exploring the synergistic effects of multiple drivers, including climate change and human activities, on savanna structure and function. Understanding these complex interactions is crucial for predicting future savanna dynamics and their implications for biodiversity, carbon storage, and ecosystem services.
Summary updated