Do limits exist on potential woody cover of Brazilian savanna?

With the increasing concern of woody plant encroachment worldwide in savanna ecosystems, many studies endeavor to examine the upper bound of woody plant cover, which is referred to as potential woody cover. Potential woody cover defines the maximum possible woody plant encroachment and bears strong...

Full description

Bibliographic Details
Main Authors: Xuebin Yang, Kelley A. Crews, Thoralf Meyer
Format: Article
Language:English
Published: Elsevier 2023-06-01
Series:Ecological Indicators
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S1470160X2300362X
Description
Summary:With the increasing concern of woody plant encroachment worldwide in savanna ecosystems, many studies endeavor to examine the upper bound of woody plant cover, which is referred to as potential woody cover. Potential woody cover defines the maximum possible woody plant encroachment and bears strong implications on savanna dynamics. Both African and North American savannas have been reported to exhibit limits on potential woody cover below 650 and 660 mm of mean annual precipitation (MAP), respectively. At present it is less clear whether that limit exists and at what level in the Brazilian savanna (the Cerrado) of South America, throughout which MAP is above 790 mm. This study models the potential woody cover pattern of Brazilian savanna over the present precipitation gradient and in relation to surface and subsurface soil moisture (SM) separately. Remotely sensed products were processed in Google Earth Engine (GEE), and included MODIS tree cover (MOD44B), TRMM monthly precipitation (3B43), and NASA-USDA Enhanced SMAP Global soil moisture. Quantile regression results suggest that below respective thresholds of MAP, surface SM, and subsurface SM, potential woody cover in Brazilian savanna is constrained. Comparison to the savanna in central Texas of the United States reveals that water availability is not the only determinant of potential woody cover. Regional context such as precipitation seasonality and woody plant species are important factors. This study also underscores the discrepancy between potential woody cover modelled based on MAP and that based on soil moisture (surface and subsurface).
ISSN:1470-160X