Estimation of Some Applied Important Parameters in Prediction of Wheat Growth

For physiological importance of parameters of nonlinear regression in sigmoidal growth patterns, eight nonlinear regression models (Beta 1, Beta 2, Logistic, Richards, Gompertz, Symmetrical expolinear, Truncated expolinear, Weibull) used for describing of changing trend of accumulated dry matter and two models as: Logistic and Beta for qualification of changing trend of LAI in two cultivar of wheat in two level of N-fertilizer. Thus, an factorial experiment in RCB design with four replications performed which treatment were two level of N-fertilizer and two cultivar of wheat (0, 150 kg/h (cv. Kohdasht) + 0, 120 kg/h (cv. Durum)) during the seasons of 2013/2014 and 2014/2015 in the research field of Gonbad Kavous University, then analyzed in combination method. Results of plotting of LAI data in logistic and beta models showed that both of models described changing trend of LAI and in either cultivar or model, applying of fertilizer decreased RMSE and upgraded R2 slowly. LAImax of cv.kohdasht was more than it in cv.durum in each fertilizer level and models. Also, results revealed that all models could describe changing trend of accumulated dry matter in either fertilizer level, but Richards, symmetrical-expolinear, Truncated-expolinear, Weibull models were some better than others. Further estimated value of parameters in these models (maximum accumulated dry matter, RGR in linear phase, RGR in expolinear phase, missed time to beginning of expolinear phase, slope of dry matter and time of CGRmax) are very practical in simulation studies, cultivars comparings, growth analyses and simulation of growth and production of wheat.