COS 73-7 - Topo-edaphic controls over woody biomass in South African savannas

Wednesday, August 10, 2011: 3:40 PM
6B, Austin Convention Center
Matt S. Colgan, Environmental Earth System Science, Stanford University, Menlo Park, CA and Gregory P. Asner, Department of Global Ecology, Carnegie Institution for Science, Stanford, CA
Background/Question/Methods

The critical role savannas play in the global carbon cycle has been recently recognized, and over the past decade much attention has been given to fire and large herbivore interactions with plant carbon pools and fluxes. However, the physical template of topography and soil provide bottom-up controls on plant biomass that interact with climate and disturbance regimes in complex ways often difficult to understand at the plot-scale, given the multiple spatial scales across which these processes occur. Here we investigate the importance of hillslope topography and soil properties in controlling woody plant aboveground biomass (AGB) using airborne LiDAR (Light Detection and Ranging) over 70,000 ha of Kruger National Park (KNP), South Africa. Applying this novel approach, we address the following: 1) Is there a catenal pattern in woody biomass on basaltic substrates, similar to that found on the granites? 2) To what extent do clayey soils on granitic footslopes mediate woody plant growth, compared to basaltic clayey soils? 3) How do regional differences in geologic parent material compare to precipitation gradients in their control over savanna woody biomass? These questions drive us towards understanding the influence of topography and soil properties, relative to climate and disturbance regimes, on the spatial heterogeneity of tree and shrub biomass in savannas.

Results/Conclusions

On basalt substrates we found a strong, negative linear relationship between mean AGB and elevation above nearest stream (R2 = 0.91), in contrast to a strong, non-linear (fourth-order) relationship on granite substrate (R2 =0.87). The latter is consistent with known qualitative catenal patterns including higher biomass on sandy crests relative to footslopes, whereas the former suggests AGB on basalt follows a more simple geomorphologic relationship, dependent on flow accumulation and water retention. AGB is more sensitive to substrate and hillslope-position than regional mean annual precipitation (MAP) gradients, likely due to high interannual variation in precipitation, where soil water storage increases survival during droughts and enables repeated resprouting in a region with a natural fire return period of four years. We also mapped AGB inside and outside herbivore/fire exclosures, and we found fire and herbivory interactions with basalt-derived soils strongly down-regulate AGB from its potential maximum, whereas on sandier soils higher hydraulic conductivity and lower grass production attenuates herbivore and fire effects. We estimate total AGB in KNP at 53.8 Tg dry mass (28.3 Mg ha-1), provide a baseline for biomass change detection due to woody encroachment, mega-herbivore impacts, and climate change.

Copyright © . All rights reserved.
Banner photo by Flickr user greg westfall.