Grass-dominated ecosystems in arid and semi-arid regions around the world have experienced increased woody plant abundance during the past century. This globally widespread phenomenon has dramatically altered the structure and function of grassland and savanna ecosystems, with the potential to profoundly influence grassland biodiversity, hydrology, biogeochemistry, and livestock production. Since arid and semi-arid systems occupy more than 40% of Earth’s land surface, geographically widespread vegetation changes from grass dominated to woody dominated ecosystems could potentially influence the global biogeochemical cycles and the climate system.

     Woody encroachment into grass-dominated ecosystems may also increase the spatial heterogeneity of soil properties, making it more difficult to accurately quantify ecosystem properties and processes based on micro-sites and limited sample size.  Previous studies have demonstrated the value of quantifying spatial patterns of soil properties and associated variability to study related processes in arid and semi-arid regions with patchy vegetation, but their results were restricted to the topsoil (0-20 cm).  Woody encroachment has been shown to substantially alter the spatial pattern and variability of soil carbon (C), nitrogen (N), and phosphorus (P) in topsoil at the individual woody patch scale from the center of the patch to the perimeter, and at the landscape scale where the landscape is comprised of different sizes of woody patches within a grassland matrix.  However, no study has explicitly assessed the extent to which woody encroachment alters the spatial pattern of C, N, and P in deeper portions of the soil profile. Non-spatial soil core studies suggest that C, N, and P pool sizes deeper in the soil profile may be modified following woody encroachment.  This implies that potentially strong spatial gradients of those elements may exist along the soil profile at the landscape scale, because organic matter inputs from litterfall and root turnover will be more concentrated in topsoil compared to deeper portions of the profile. Given the fact that most studies on spatial patterns of SOC following woody encroachment have been conducted on surface soils, and that substantial C, N, and P are stored in subsurface soils, this represents a major knowledge gap concerning the effects of vegetation cover change on the pattern of spatial heterogeneity of C, N, and P along the entire soil profile.    

     The primary purpose of this study was to determine how woody encroachment into grassland affects the direction, magnitude and pattern of spatial heterogeneity in C, N, and P along the soil profile. To accomplish this, 320 randomly located and spatially-specific soil samples were taken to a depth of 1.2 m across a 160 m x 100 m landscape in a subtropical savanna ecosystem where woody plants (dominated by mesquite, Prosopis glandulosa) have been encroaching into grasslands during the past century.  The study area was situated in the Rio Grande Plains region at the Texas A&M AgriLife La Copita Research Area located 65 km west of Corpus Christi, Texas, USA.  Soil samples were analyzed for root biomass, soil organic carbon (SOC), total nitrogen (TN), and total phosphorus (TP).

     Woody encroachment increased SOC, TN, and TP storage throughout the entire 1.2 m soil profile, albeit at different rates. SOC and TN accumulations were coupled following woody plant encroachment; that is, both elements increased proportionally, and had similar spatial patterns throughout the soil profile that were strongly linked to spatial patterns of grassland vs. woody patches. Spatial patterns of TP also strongly resembled those of vegetation cover throughout the soil profile.  However, TP increased more slowly than SOC and TN in surface soils (0-5 cm) but faster in subsurface soils (15-120 cm).  Proportionally more P accrued in deeper portions of the soil profile beneath woody patches where alkaline soil pH and high carbonate concentrations would favor precipitation of P as relatively insoluble calcium phosphates. This imbalanced relationship highlights that the relative importance of biotic vs. abiotic mechanisms controlling C and N vs. P accumulation following vegetation change may vary with depth.

     Land cover/land use changes often perturb soil biogeochemistry and are considered essential components of coupled biogeochemistry and climate models. Recent studies have recognized the importance of incorporating P cycling into coupled climate-carbon cycling models and C-N-P interactions into ecosystem models. Our findings suggest that modelers should be aware of the fact that the biogeochemical controls over C and N vs P cycling following vegetation change may be different, and decoupling of P cycling from C and N cycling may lead to differential responses to future perturbations and changes. Specific drivers and mechanisms of P cycling should be captured in model development, especially when subsurface soils are considered.  This study highlighting the complex responses of soil C, N, and P to vegetation change has far reaching implication for future empirical and modeling studies aimed at understanding the biogeochemistry of dryland ecosystems which are particularly fragile with respect to anticipated global changes.

Last Modified: 05/25/2018
Modified by: Thomas W Boutton