Wheat blast is a devastating disease caused by the Triticum pathotype of Magnaporthe oryzae (MoT). Studies have shown that MoT inoculum and blast symptoms can spread from leaves to spikes, suggesting that wheat blast might follow a polycyclic pattern. There is increasing need to study the temporal dynamics of MoT and wheat spike blast (SB). In addition, early detection and accurate identification of SB are critical for implementing adequate management strategies. Two field experiments were conducted in Bolivia to characterize the change in SB severity and incidence over time and determine if multispectral imagery can be used to accurately assess SB intensity. Disease progress curves and regression models were fitted to describe SB epidemics. The logistic population growth model best described SB incidence and severity change over time in both inoculated and non-inoculated plots from both locations (R²=0.87-0.98; MSE=0.24-0.94). Lin’s concordance correlation coefficients were estimated to measure agreement between visual and digital estimates of SB intensity and to estimate accuracy, precision, and bias. Multispectral aerial imagery accurately (?_c = 0.65-0.91) and precisely (r = 0.1-0.95) estimated SB incidence and severity in inoculated and non-inoculated plots at both locations. Our findings suggest that SB might follow a polycyclic pattern and that SB incidence and severity can be quantified using aerial imagery. Higher resolution imagery of plants showing initial SB symptoms would likely increase electronic sensor-based disease detection and quantification in the field.