Global Estimation of Soil Nitrous Oxide Emission Using a Semi-Empirical Model and a Global Dataset

Abstract:

Nitrous oxide (N₂O) flux is one of the major greenhouse gas fluxes from terrestrial ecosystems. N₂O is produced as a by-product of the transformations of nitrogen in the soil and is generally released from the soil surface to the atmosphere. The purpose of this study is to provide a data-oriented global estimate of N₂O flux from the soil. In this study, a semi-empirical model was developed by modifying a model for soil CO₂ flux (Raich et al. 2002). The model consists of the functions of nitrogen, air temperature, and precipitation. The parameters of the functions were determined using a global dataset of N₂O emission (Stehfest and Bouwman 2006) through the Bayesian approach. The function of nitrogen is a function of carbon to nitrogen ratio. For agricultural land use, the effect of nitrogen fertilizer was also incorporated in the function of nitrogen. The model was applied at a spatial resolution of 5 min and at a monthly time resolution.

Preliminary calculation revealed that the total amount of N₂O emission was 10 Tg N yr⁻¹. The upland natural land was the major source of N₂O emission, followed by the upland agricultural land. Latitudinally, the flux was high at 30°N–40°N and 10°S–10°N. The major contribution to the peak at 30°N–40°N was by the upland agricultural land and that to the peak around the equator was by the upland natural land. The monthly flux showed a clear seasonality, and it was the highest and lowest in August and February, respectively. The global scale seasonality was mainly contributed by the N₂O flux from the upland agricultural land in the northern mid-latitude. This study provides data-oriented spatiotemporal distribution of soil N₂O flux, and I hope that these data will be used for a benchmark and constraint of process-based modeling.