Field experiments were conducted in Kentucky and Louisiana in 2008 and 2009 (split-plot in a randomized complete block design with four replications) to investigate it is possible to simulate with manual defoliation the effect of soybean rust (SBR) (Phakopsora pachyrhizi Syd. and P. Syd) injury on a healthy soybean [Glycine max, (L.) Merr.] canopy, understand how defoliation affects the growth dynamics and canopy light interception, and if defoliation affectsleaf senescence and nitrogen remobilization during the seed-filling period. Two manual defoliation treatments based on changes in effective leaf area index (ELAI) (calculated as the reduction in leaf area equivalent to SBR-induced premature leaf abscission, loss in green leaf area, and reduction in photosynthetic capacity of diseased leaves) in infected canopies in Brazil were used to simulate SBR infection at growth stage R2 (full flowering) and R5 (beginning of seed-fill). Both defoliation treatments reduced yield in all experiments and the reduction was larger for the treatments at growth stage R2. The yield losses were equivalent to that observed in infected soybean canopies in Brazil. This suggests that a system of manual defoliation to simulate changes in effective leaf area duration shows promise as a tool to simulate the impact of SBR on soybean yield. The radiation use efficiency and crop growth rate from growth stage R2 to R5 were not influenced by defoliation. Defoliation started at growth stage R2 reduced seed number per unit area, while defoliation started at growth stage R5 reduced seed size due to shortening the seed-fill duration and a lower seed growth rate. There is no evidence that manual defoliation affected leaf senescence or nitrogen redistribution to the seed. This study found that the reduction of light interception by SBR was the main reason for the reductions in soybean growth and yield.