TITLE:
Quantifying and Validating Soybean Seed Emergence Model as a Function of Temperature
AUTHORS:
Firas Ahmed Alsajri, Chathurika Wijewardana, L. Jason Krutz, J. Trenton Irby, Bobby Golden, K. Raja Reddy
KEYWORDS:
Growing Degree Days Model, Seed Emergence, Soybean, Temperature
JOURNAL NAME:
American Journal of Plant Sciences,
Vol.10 No.1,
January
18,
2019
ABSTRACT: Developing
a model for soybean seed emergence offers a tool producers could use for
planting date options and in predicting seedling emergence. In this study,
temperature effects on soybean seed emergence were quantified, modeled, and
validated. The data for seed emergence model development was generated at
varying temperatures, 20°C/12°C, 25°C/17°C, 30°C/22°C, 35°C/27°C, and
40°C/32°C, on
two soybean cultivars, Asgrow AG5332 and Progeny P 5333 RY. Time for 50%
emergence (t50%) was
recorded, and seed emergence rate (SER) was estimated as reciprocal to time at
each temperature in both the cultivars. No differences were observed between
the cultivars in their response to temperature. A quadratic model (QM) best
described the relationship between t50% and SGR and temperature
(R2 = 0.93). Two sets of experiments were conducted to validate the
model. In Experiment 1, 17 time-series planting date studies with the same
cultivars were used by utilizing diurnal and seasonal changes in temperature
conditions. In the second experiment, sunlit growth chambers with 3 different
day/night temperatures, low—20°C/12°C, optimum—30°C/22°C, and
high—40°C/32°C, and
64 soybean cultivars belonging MG
III, IV, and V, were used. Air temperature and t50 were recorded, and SGR was
estimated in all experiments. No differences were recorded among the cultivars
for t50% and
SGR, but differences were observed among seeding date and temperature
experiments. We tested QM and traditionally used Growing Degree Days models against the data
collected in validation experiments. Both the model simulations predictions
agreed closely with the observed data. Based on model statistics, R2,
root mean square errors (RMSE), and comparison of observations and predictions
to assess model performance, the QM model performed better than the GDD model
for soybean seed emergence under a wide range of cultivars and environmental
conditions.