AbstractLate watergrass is a competitive weed of rice that is well adapted to both aerobic and anaerobic environments. Cultural controls such as a stale-seedbed and alternating from wet- to dry-seeding have been proposed as management options. However, the effects of these systems on its emergence and early growth are unknown. The objective of this study was to modify a previously developed population-based threshold model (PBTM) to predict emergence and early growth under field conditions. In 2013, a series of experiments were conducted at the California Rice Experiment Station (CRES) in Biggs, CA, to evaluate emergence and early growth of multiple herbicide–resistant and -susceptible late watergrass at four burial depths (0.5, 2, 4, and 6 cm) under three irrigation regimes: continuously flooded (CF), daily flush (DF), and intermittent flush (IF). Resistant plants emerged at a significantly higher rate under the IF treatment (P < 0.05). Both biotypes showed decreasing emergence with increasing depth, and no plants emerged from the 4- or 6-cm depths in the CF treatment. Using the Gompertz growth curve, resistant plants had greater predicted growth rates (k), lower predicted maximum heights (hmax), and a shorter time to predicted maximum growth rate (tm) than susceptible plants under the CF and DF treatments. Under the IF treatment, the susceptible plants had greater k, lower hmax, and shorter time to predicted tm. Information about burial depth and irrigation was incorporated into a previously developed PBTM for late watergrass, and validated at the CRES in a field with a susceptible late watergrass population in 2013 and 2014, under two irrigation systems, CF and IF. Model fit was best in the CF treatments (average Akaike information criteria [AIC] = 199.05) compared to the IF treatments (average AIC = 208.6).