Crop Model Application to Analyze the Effects of Long-Term Changes in Climate and Soil Fertility on Rice Yield – a Case Study with a 47-Year Experiment with NPK and Compost Applications to Paddy Fields.

Evidence is increasing that recent changes in climate have started to influence various aspects of crop production. The size of the effects is important to predict future influence of the climate and develop technology under climate change. Long-term field trials are a valuable source of information to analyze the effects of long-term change in climate. However, multiple climatic factors such as atmospheric CO₂ concentrations are concurrently rising. Changes in climate can also affect soil fertility, which can indirectly influence crop production. Consequently, the effect of single factor is often difficult to determine with a simple statistical analysis. Here, we attempt to apply rice growth model (H/H) to disentangle these factors on rice yield. We used the data from long-term field trials, initiated in 1967 at Daisen city, Akita, Northern Japan (39°32'50"N 140°22'34"E), with different treatments of continuous application of NPK, and/or compost application to the paddy fields of gray lowland soil (Gleyic Fluvisols). A japonica rice cultivar “Kiyonishiki” was grown all except for the first six years. The fields were fully irrigated (submerged) during the growing season, with occasional surface water drainage. H/H model simulates rice growth and yield based on daily climatic factors such as CO₂, temperature, solar radiation, relative humidity and soil factors such as nitrogen supply from the soil organic matter and fertilizer. During this period, a significant increase was observed in air temperature by about 1.4 °C and in CO₂ by about 76 ppm. The model was first run with the actual practice and climate data, then with the detrended climatic variables. By comparing the simulated results with different climatic and soil conditions, we estimated the effects of each factor on rice yield. Overall, a significant positive trend was observed in the observed yields for all treatments, ranging from 1.5 to 4.6 gm^-2yr^-1, depending on the amount of compost applied. The yield trend associated with climate averaged about 0.7 gm^-2yr^-1, as a balance between opposing effects of positive effects of CO₂ (1.5 gm^-2yr^-1) and temperature (-0.8 gm^-2yr^-1).