RESUMO Os coeficientes genéticos precisam ser calibrados para as cultivares mais representativas do mercado, portanto, é necessário adicionar novas cultivares a modelos como o SimulArroz. Portanto, o objetivo deste estudo foi calibrar e avaliar o modelo SimulArroz para duas novas cultivares de arroz irrigado. Os experimentos foram conduzidos no município de Goianira em Goiás durante quatro safras (2014/15, 2015/16, 2016/17, 2017/18) e no Rio Grande do Sul nos municípios de Alegrete (2015/16), Cachoeirinha (2015/16), Capão do Leão (2016/17, 2017/18), Santa Vitória do Palmar (2017/18) e Uruguaiana (2014/15, 2015/16). O delineamento utilizado foi o de blocos casualizados, com quatro repetições no Rio Grande do Sul e com parcelas de semeadura em Goianira. Foram utilizadas as cultivares BRS Catiana e BRS Pampa, sendo avaliadas o estádio Haun (HS), fenologia, biomassa de matéria seca da parte aérea e produtividade. A raiz do erro quadrático médio (RMSE) para matéria seca aérea variou de 51,7 a 577 g m-2, e para produtividade, a raiz do erro quadrático médio normalizado (NRMSE) variou de 24 a 32% e 22 a 35% para o potencial e alto nível tecnológico, respectivamente. O modelo SimulArroz foi capaz de prever satisfatoriamente o crescimento, desenvolvimento e produtividade para as cultivares BRS Catiana e BRS Pampa, aumentando sua área de aplicação e abrangendo também a região tropical do Brasil. mercado portanto Portanto irrigado 2014/15, 201415 2014 15 (2014/15 201516 2015 16 2015/16 201617 2016 17 2016/17 2017/18 201718 2017 18 , (2015/16) (2016/17 2017/18, (2017/18 2015/16. . 2015/16) casualizados Pampa HS, HS (HS) fenologia RMSE (RMSE 517 51 7 51, 57 m2, m2 m 2, 2 m-2 NRMSE (NRMSE 32 35 tecnológico respectivamente crescimento Brasil 2014/15 20141 201 1 (2014/1 20151 2015/1 20161 2016/1 2017/1 20171 (2015/16 (2016/1 (2017/1 (HS 5 m- 3 2014/1 20 (2014/ 2015/ 2016/ 2017/ (2015/1 (2016/ (2017/ 2014/ (2014 (2015/ (2016 (2017 (201 (2015 (20 (2 (
ABSTRACT Given genetical coefficients need to be calibrated for the most important cultivars on the market, new cultivars must be added to models such as SimulArroz. Thus, the aim of this study was to calibrate and evaluate the SimulArroz model for two new irrigated rice cultivars. The experiments were conducted in the municipality of Goianira in Goiás state during four growing seasons (2014/15, 2015/16, 2016/17, 2017/18) and in Rio Grande do Sul state in the municipalities of Alegrete (2015/16), Cachoeirinha (2015/ 16), Capão do Leão (2016/17, 2017/18), Santa Vitória do Palmar (2017/18) and Uruguaiana (2014/15, 2015/16). A randomized block design was used, with four replicates in Rio Grande do Sul and sowing plots in Goianira. The BRS Catiana and BRS Pampa cultivars were used and the Haun stage (HS), phenology, shoot dry matter biomass and yield were evaluated. The root mean square error (RMSE) for above-ground dry matter ranged from 51.7 to 577 g m-2, and for yield, the normalized root mean square error (NRMSE) ranged from 24 to 32% and 22 to 35% for the potential and high technological levels, respectively. The SimulArroz model was able to satisfactorily predict the growth, development, and yield of the BRS Catiana and BRS Pampa cultivars, increasing their area of application, including the tropical region of Brazil. market Thus 2014/15, 201415 2014 15 (2014/15 201516 2015 16 2015/16 201617 2016 17 2016/17 2017/18 201718 2017 18 , (2015/16) 2015/ (2015 16, 16) (2016/17 2017/18, (2017/18 2015/16. . 2015/16) HS, HS (HS) phenology evaluated RMSE (RMSE aboveground above ground 517 51 7 51. 57 m2, m2 m 2, 2 m-2 NRMSE (NRMSE 32 35 levels respectively growth development application Brazil 2014/15 20141 201 1 (2014/1 20151 2015/1 20161 2016/1 2017/1 20171 (2015/16 (201 (2016/1 (2017/1 (HS 5 m- 3 2014/1 20 (2014/ 2016/ 2017/ (2015/1 (20 (2016/ (2017/ 2014/ (2014 (2 (2016 (2017 (