COS 10-9 - Tree species modification of soil moisture, and soil moisture influence on soil carbon dioxide efflux, in tropical forest plantations

Monday, August 7, 2017: 4:20 PM
D132, Oregon Convention Center
James W. Raich, Steven J. Hall and Chaoqun Lu, Ecology, Evolution & Organismal Biology, Iowa State University, Ames, IA
Background/Question/Methods

Evergreen tropical rainforests cover only about 8% of Earth’s land surface, yet they strongly influence the global carbon cycle as a result of their ever-warm and ever-moist conditions, which promote rapid biological activity year-round. They also are diverse, biologically and structurally, and that diversity generates uncertainty in the extrapolation of observations from individual sites to larger spatial scales. For example, among evergreen tropical and subtropical moist forests, soil-CO2 emissions average 1600 (SE 46) g m-2 year-1 of carbon. However, within six types of replicated, monodominant tree plantations in lowland Costa Rica having the same land-use histories, soil types, and climates through time, mean soil-CO2 emissions ranged from 1550 to 2010 g m-2 year-1. Individual tree species had strong and consistent influences on soil carbon dioxide production. They also modified the physical environments of their plots. It is widely recognized that species respond to their environments, but also modify them. Similarly, soil CO2 emissions vary in response to forcing factors such as temperature and moisture, but also with biological factors such as plant productivity. Recognizing these dualing paradigms - biotic and abiotic - for explaining observed differences in soil-CO2emissions among sites, we first evaluated whether differences in carbon capture and allocation among tree species could explain observed differences in soil CO2fluxes and, second, quantitatively assessed the importance of species modulation of soil physical properties.

Results/Conclusions

Across 16 plots of four tree species in lowland Costa Rica, there was no relationship between mean annual soil CO2 flux and aboveground litter production. There was, however, a weak relationship (P < 0.10) between soil CO2 emissions and aboveground net primary productivity (ANPP), suggesting that total carbon flux plays a discernible role in defining the magnitude of soil respiration among forest types at a single location. Tree species significantly altered surface soil water contents in this rainy environment (P < 0.0001), with the driest soils (mean 42 % of total soil volume) occurring in plantations of Pentaclethra, and the wettest soils being in plantations of Virola, where soil moisture averaged 47% of total soil volume. Strongly non-linear relationships between surface-soil water content and soil CO2 emissions were apparent, with maximum emissions occurring when soil water contents were <40%. Virola, in particular, appears to promote soil saturation, but soil CO2 emissions in all four plantation types seem to be strongly limited by too much water.