Crop seed germination is a critical factor in food production. Germination traits vary between different vegetables and between cultivars of the same vegetable. Traits such as germination proportion and the rate and uniformity of germination also contribute to the success of vegetable cultivars to regional temperature averages and regimes. Temperature is a major abiotic factor in seed germination, and selection of seed varieties suited for local temperatures is important to successful crop establishment. Here we aim to model the germination traits of several Korean and North American vegetable cultivars in relationship to temperature. We hypothesize that a model based on time and temperature will represent germination traits across multiple cultivars and species of vegetable. Additionally, we hypothesize that predicted germination traits will be similar within species but will vary between cultivars of the same species based on the regional temperature norms. Eleven vegetable cultivars including two corn, four radish, two pepper, and three onion, were germinated in growth chambers set at 5°C intervals between 5 and 40°C with a 12 hour photoperiod. Germination, defined by radicle emergence equal to the length of the seed, was monitored and recorded daily. Two modeling approaches were used. 1) A time to event model using the drcSeedGerm package in R was used to determine maximum germination proportion (Pmax), germination rate at 50% germination (GR50), and uniformity. 2) A 12-parameter compartmental temperature and time model was implemented, parameterized, and validated in the Cropbox modeling framework. Results show differing temperature responses in germination traits. Additionally, high uniformity was observed in most varieties within the optimal temperature range. The optimal temperature for germination was broad for the vegetable cultivars tested here, with a rapid decrease in Pmax at high and low temperature extremes, except in onions, which showed high Pmax even at 5°C, and a gradual decline above 25°C. All cultivars showed decreased Pmax at 40°C. A time to event model was able to predict germination traits in several crop species. The compartmental model was better equipped to handle heat induced seed degradation but was less parsimonious for determining germination probability and germination rate related parameters. Modeling crop germination traits can provide important context for selecting appropriate cultivars for local climates. A thermal time to event model and a compartmental model both provide potential frameworks for modeling germination traits of diverse vegetable species.
