The air temperature effect on plant development can be represented using the thermal time approach. In order to calculate the thermal time, the base temperature, below when development stops or takes place at very low rates, is needed. This work was carried out to estimate the base temperature for node appearance and compared dates for cultivation for variable plastochron in watermelon crop. Three field experiments were conducted in Santa Maria, RS, Brazil, with eight sowing dates during the 2006-2007 growing season (05/09/2006 and 21/09/2006), 2008/2009 (20/09/2008, 06/10/2008, and 20/10/2008) and 2009/2010 (20/09/2009, 21/10/2009, and 30/11/2009). The cultivar "Crimpson Sweet" was used in a randomized block design with four replications per sowing dates. The accumulated number of nodes on the main stem (NN) of watermelon was observed in three plants, three times a week. Plastochron was estimated by the inverse of the slope in the linear regression of NN against accumulated thermal time. The temperature base estimated using the least Mean Square Error (MSE) approach was 7.0 °C. The plastochron in watermelon was not different among sowing dates during the three growing seasons, with an average of 23.4 ºC day node-1. From a modeling perspective, the implications of these results are that a single plastochron value can be used to simulate node appearance of the "Crimpson Sweet" cultivar, regardless of the sowing date.
Citrullus lanatus; Vegetative development; Thermal time
