Modelling seedling emergence in Paspalum species using environmental data from field experiments

In warm-temperate regions, the adoption of warm-season forage grasses has been hindered by low and unpredictable seedling emergence because of seed dormancy. Developing models driven by temperature (T) and soil water potential (psi) may provide tools for choosing adequate sowing conditions. Models are usually developed with controlled chamber germination data, but they can be built using field emergence and weather station data. Seedling emergence data of four Paspalum populations (two dallisgrass subspecies, bahiagrass and vaseygrass) were gathered from five experiments in three Campos locations (Buenos Aires, Montevideo and Salto) during two years, with fall and spring sowing dates and irrigated and non-irrigated plots. Thermal and hydrothermal time models were used for emergence timing. The predictions were better for irrigated plots and fall sowings. For maximum emergence proportion (MEP), a mixed linear model was adjusted. Within a non-irrigated treatment, the higher MEP was for early-spring sowings, but the lowest for late-spring ones. Additionally, a thermal time index (TTI) was modified to model MEP. A coefficient which weighs degree-days accumulation according to the hydrothermal range (HTR) of each day was set. The HTRs were defined by T and psi thresholds. Thresholds and coefficient values were optimized by genotype until linear regressions between MEP and modified TTI achieved higher fit. Days with high temperature (>18 degrees C or 20 degrees C according to genotype) accumulated half or none of degree-days when high- or mid-soil water content was available, respectively. Differences among Paspalum genotypes, sowing date recommendations and modified TTI usefulness were discussed.

Automated summary

In this study, models driven by temperature and soil water potential were developed to predict seedling emergence for four Paspalum populations under different sowing conditions. Results showed that irrigated plots and fall sowings had better emergence predictions. Adjusting maximum emergence proportion and thermal time index can improve the accuracy of emergence prediction.