Author ORCID Identifier

Year of Publication


Degree Name

Doctor of Philosophy (PhD)

Document Type

Doctoral Dissertation


Agriculture, Food and Environment


Plant and Soil Sciences

First Advisor

Dr. Mary A. Arthur


Tropical forest soils contain one-third of global soil carbon (C). The warm and moist climate in tropical forests leads to rapid soil organic carbon (SOC) decomposition, with the highest soil microbial respiration rates in the world, so even a slight change in soil C and microbial respiration could affect atmospheric carbon dioxide concentration. However, there remains a lack of understanding of the mechanisms driving microbial respiration in tropical forests, due to different climate and biophysical drivers compared to temperate or boreal forests. Furthermore, forest conversions (from natural forests to plantations) are most widespread in tropical regions, leading to a loss of SOC during harvesting and reforestation processes. However, the magnitude and direction of SOC changes vary greatly. These uncertainties result in great variability in, and disagreement among, models predicting the response of SOC to future climate in tropical forests.

To quantify the contribution of recent photosynthesis and SOC decomposition to soil microbial respiration, I separated microbial respiration from total soil respiration using the trenching method with 150µm pore size of nylon mesh sheets in a subtropical moist evergreen broadleaved (SMEB) forests. In addition, I excavated five 50cm x 50cm mineral soil columns from the forest and incubated them in the open. Measures of soil microbial respiration, soil temperature and moisture, and photosynthetically active radiation (PAR) were automated, recorded simultaneously every 30 minutes. Results demonstrated that current photosynthates contributed 88% of C sources to soil microbial respiration, and soil temperature was not the controlling factor for Rm on the diel scale; rather, soil microbial respiration was strongly controlled by PAR. These results suggest the need for a new conceptual model of C cycling in SMEB forests, in contrast to most microbial respiration models based on temperature-dependent SOC decomposition in temperate forests.

The intensification of human activities in forest management have led to large SOC losses during the establishment of tree plantations from former natural forests. I quantified three major SOC loss pathways in the establishment of a Chinese fir plantation, demonstrating that converting of natural forest to Chinese fir plantation resulted in 28% loss of SOC present prior to conversion - 11% through volatilization by slash burning, 10% via soil erosion, and 7% via increasing SOC mineralization. Forest conversion also altered the available C sources for soil microbes from newly formed carbohydrates in the natural forest to SOC in the plantation. Finally, I found SOC did not recover after 40 years. My results highlight that slash burning during forest conversion leads to a dramatic, and lasting, decline in SOC in SMEB forests during the establishment of tree plantations converted from former natural forest.

Although we have a good understanding of the mass loss rate and nutrient release of plant residues during decomposition, the role of incorporation of plant residues into SOC is still poorly understood and often neglects the influence of live roots on SOC sequestration. In this study, I examined changes in bulk and rhizosphere SOC across three developmental stages of Chinese fir plantation (6, 18, and 42 years old) with age-associated differences in net primary productivity (NPP). My results indicated that SOC concentration in bulk soils did not vary significantly with forest age, but SOC concentration in rhizosphere soils was higher in 6 yrs and 42 yrs stands, both of which had lower NPP. Both labile and recalcitrant C pools in rhizosphere soils were smallest in the 18 yrs stand, which had the highest NPP. Variation in rhizosphere SOC was correlated with soil phosphorus and microbial biomass carbon: nitrogen ratio (MBC:MBN), which drove labile and recalcitrant SOC dynamics in different ways depending on forest age. My results demonstrated that variability in forest productivity and nutrient demand with time are key factors controlling rhizosphere SOC sequestration and nutrient cycling.

Altogether, this work provides insights into the magnitude and pathways of SOC losses due to conversion of natural forest to tree plantation in SMEB forests, and its feedback to tree plantation development in a rotation. This study also highlights the dominance of photosynthates on soil microbial activity and SOC sequestration. The findings from these studies will improve our understanding of C cycling in tropical forests and reduce the uncertainty in modeling future C dynamics in tropical forests.

Digital Object Identifier (DOI)

Funding Information

This research was supported by the National Natural Science Foundation of China of 31670623 from 2017-2020 and 31930071 from 2020-2024, respectively.

Available for download on Tuesday, April 27, 2021