Date of Award

Spring 2000

Project Type


Program or Major

Natural Resources

Degree Name

Doctor of Philosophy

First Advisor

Patrick Crill


Soils are the major natural source of nitrous oxide (N2O). Intensive land use increased atmospheric concentrations of this greenhouse gas. Soil microbes produce and consume nitrogen oxides (NO, N2O) during the processes of nitrification (aerobic) and denitrification (anaerobic). Micro-scale variability of controlling factors cause nitrification and denitrification to occur simultaneously in soils, resulting in high spatial variability of nitrogen oxide emissions. Fertilization increases nutrient availability and thus N2O fluxes. Forest soils in the humid tropics account for 20--50% of all N2O sources. Expansion and intensification of tropical agriculture is expected to increase atmospheric N2O concentrations.

We measured N2O fluxes from secondary humid tropical forest soils in Costa Rica, followed fluxes during forest conversion and studied emissions from unfertilized and fertilized agricultural soils. We related fluxes to soil moisture dynamics and agricultural practice. Gases were measured using manual and automated chamber techniques. Soil moisture content was measured using manual (auger) and automated (Time Domain Reflectrometry; TDR) sampling techniques. A 3-phase-mixing model was found suitable to calibrate TDR technique for the studied soils. The field experiment was based on a split-plot design, comparing clay versus loam, each under fertilized and unfertilized annual and perennial crop. Soils feature relatively low bulk density, high hydraulic conductivity and high organic matter content. Mean soil moisture content was above 70% water-filled-pore-space (WFPS) in both soils and land uses. N 2O was emitted throughout the year.

Fluxes from forest soils showed no seasonality. Forest conversion caused fluxes to increase temporarily. Fertilization was the dominant source for temporal variability under agricultural use, differences in flux dynamics were large between individual post-fertilization phases. N2O-loss as % of applied fertilizer-N increased with soil moisture. Spatial variability was generally high, especially post-fertilization. Repeated fertilization increased mean and variation of fluxes. Emissions simulated by regression models and by the physically based Denitrification-Decomposition model matched field measured fluxes well. Both modeling techniques confirmed that nutrient availability and soil moisture content were the dominant flux controls. In aggregated soils differences in soil structure between the surface layer and soil at 0.05 m depth may affect moisture content and consequently soil N 2O fluxes.