Measures to models of moisture effects on soil respiration in Panamanian lowland forests

Daniela Cusack and Lee H. Dietterich, Ecosystem Science & Sustainability, Colorado State University, Fort Collins, CO, Daniela Cusack, Smithsonian Tropical Research Institute, Ancon, Panama, Benjamin N. Sulman, Climate Change Science Institute and Environmental Sciences Division, Oak Ridge National Laboratory, Oak Ridge, TN

Background/Question/Methods
Humid tropical forests contain some of the largest soil carbon (C) stocks on Earth, which are facing projected ecosystem-scale drying and drought. Drying in these ecosystems is likely to change soil C losses via effects on soil respiration. In particular, drying in the context of fluctuating redox during wet periods, and moisture limitation during dry periods, may stimulate microbial respiration in some sites and suppress it in others depending on baseline soil moisture levels. To investigate this, we collected monthly-scale surface soil moisture (0-10 cm), temperature, and carbon dioxide surface fluxes across four distinct lowland seasonal tropical forests that vary in background rainfall (2325-3300 mm annual precipitation) and soil fertility. We compared two years of field data against microbial-explicit soil C cycling model predictions of soil respiration rates using the Carbon, Organisms, Rhizosphere and Protection in the Soil Environment (CORPSE) model. We adjusted the moisture function in the model using field data to improve simulated moisture sensitivity of soil respiration, and then compared against the original model. Using the updated model, we ran three model scenarios, including 15% and 30% chronic rainfall reduction during the rainy season, and an episodic drought extending the dry season by one month.
Results/Conclusions
Field measurements show the lowest soil respiration during the dry season, and a peak in respiration during the early wet season before soil moisture peaked. Using the field data, we revised the model optimal moisture for maximum soil respiration from 71% to 38 – 46% of soil saturation, with different optima for different sites. Revised model outputs better matched field observations, and flipped the direction of the projected effect of some drying scenarios. Before optimization, all drying scenarios produced annual reductions in soil respiration. With the newly optimized model, all drying scenarios increased annual soil respiration for the wettest site, with the largest increase at 30% rainfall reduction. Similarly, chronic drying increased annual soil respiration for the most fertile soil. In contrast, 30% rainfall reduction reduced annual soil respiration for two drier sites on infertile soils. The extended drought scenario had the smallest effect on soil respiration, with slight increases for the wettest site only. This study illustrates the importance of optimizing model moisture functions against ecosystem-scale field data. Model predictions suggest that chronic drying is likely to have a stronger effect than episodic drought on tropical forest soil respiration, with the net effect related to background rainfall and soil fertility.