Plastic temperature response function accurately simulates crop flowering or heading date. (29th June 2020)

• Record Type: Journal Article
• Authors: Wu, Dingrong; Huo, Zhiguo; Wang, Peijuan; Song, Yanling; Yang, Jianying; Zhao, Huarong; Yang, Yang; Yu, Qiang
• Is Part Of: Agronomy Journal. Volume 112:Number 5(2020)
• Page Start: 3832
• Page End: 3846
• Publication Date: 2020-06-29
• Subjects: Agronomy -- Periodicals; 630
• DOI: 10.1002/agj2.20112
• Languages: English
• ISSNs: 0002-1962

Abstract

Abstract: Although crop phenology is responsive and adaptable to cultural and climatic conditions, many phenology models are too sensitive to variable climatic conditions. We developed a plastic temperature response function by assuming that development rate was linearly related to temperature and that the linearity was linearly responsive to day of year (DOYv ) of the starting date of the vegetative growth period (VGP). Phenology observations and weather data were acquired for winter wheat ( Triticum aestivum L.), rice ( Oryza sativa L.), maize ( Zea mays L.), and soybean [ Glycine max (L.) Merr.] at 12 locations over 15–26 yr. Additional data were observed for maize grown in an interval planting experiment. For 78.6% of the sites, the crop development rate during the VGP was positively affected by DOYv . Partial correlation analysis (controlling for temperature) indicated that DOYv was independent of temperature. When averaged over all crops and sites, the RMSE for a plastic phenology model based on both response and adaptation mechanisms was lower (RMSE = 2.81 d) than models (RMSE = 3.39) based only on response mechanism ( p < .01). Furthermore, simulations produced by the plastic model showed less bias to DOYv, temperature, and year. The plastic function provided a simple and effective method for achieving better phenology simulation accuracy. According to the plastic function, growing season under warming conditions will not be reduced by as much as simulated by models based only on response mechanism, so yield loss due to warming is likely to be overestimated.