Calibration of Daycent biogeochemical model for rice paddies in three agro-ecological zones in Peninsular India to optimize cropping practices and predict GHG emissions

Abstract:

Rice is a staple for nearly half of the world population with irrigated and rainfed lowland rice accounting for about 80% of the worldwide harvested rice area. Increased atmospheric CO₂ and rising temperatures are expected to adversely affect rice yields by the end of the 21^st century. In addition, different crop management practices affect methane and nitrous oxide emissions from rice paddies antagonistically warranting a review of crop management practices such that farmers can adapt to the changing climate and also help mitigate climate change.

The Daily DayCent is a biogeochemical model that operates on a daily time step, driven by four ecological drivers, i.e. climate, soil, vegetation, and management practices. The model is widely used to simulate daily fluxes of various gases, plant productivity, nutrient availability, and other ecosystem parameters in response to changes in land management and climate.

We employed the DayCent model as a tool to optimize rice cropping practices in Peninsular India so as to develop a set of farming recommendations to ensure a triple win (i.e. higher yield, higher profit and lower GHG emissions). We applied the model to simulate both N₂O and CH₄ emissions, and crop yields from four rice paddies in three different agro-ecological zones under different management practices, and compared them with measured GHG and yield data from these plots. We found that, like all process based models, the biggest constraint in using the model was input data acquisition. Lack of accurate documentation of historic land use and management practices, missing historical daily weather data, and difficulty in obtaining digital records of soil and crop/vegetation parameters related to our experimental plots came in the way of our execution of this model. We will discuss utilization of estimates based on available literature, or knowledge-based values in lieu of missing measured parameters in our simulations with DayCent which could prove to be a solution to overcome data limitations in modeling with DayCent and other process based models for developing regions of the world.