Comparison between logistic and Gompertz equations for predicting groundnut rust epidemic

Stepwise multiple regression analysis was used to identify those meteorological variables useful in explaining variation in groundnut rust (Puccinia arachidis) severities. Disease severity estimates (Y), maximum temperature (X1), minimum temperature(X2), maximum relative humidity (X3), minimum relative humidity (X4), total rainfall (X5) and wind velocity (X6) were used as variables. Maximum temperature, minimum temperature, maximum relative humidity and minimum relative humidity were the variablesthat had high co-relations with the disease in all three (rainy, winter and summer) seasons. Gompertz equation was best linearized with the disease progress data followed by logistic and untransformed data sets. The linear prediction equations are (1) Y=-13.1129 + 0.6698 X1 -0.3079 X2+0.0329 X3 -0.0832 X4 + 0.0262 X5 -0.1276 X6 for rainy season. For winter and summer seasons equations were (2) Y=15.1442 +0.3629 X1 -0.2205 X2 -0.3074 X3+0.0437 X4 -0.4902 X5+0.0712 X6 and (3) Y=3.2286 -0.1490 X1 + 0.131X2 -0.0411 X3 + 0.0190 X4 + 0.0330 X5 + 0.0171 X6, respectively.