In the fall of 2005, the University of Kentucky (UK) was awarded a grant (Project PI: Saratha Kumudini) to develop a soybean rust (SBR) yield loss prediction tool. The goal of the project was to develop a risk management tool to help soybean producers decide how to better manage SBR. This interactive website is the result of this collaborative effort. See the “Research Partners” page for more information about the institutions and people involved.

The first step was to better understand how SBR reduced soybean yields. Experiments were conducted in Londrina, Brazil where severe SBR epidemics often occur. SBR affects soybean yield by three main mechanisms:

1. Accelerated leaf drop.
2. Reduced green leaf area, due to necrotic/chlorotic lesions, on remaining leaves.
3. Reduced photosynthetic capacity of remaining green leaf area.

These factors combine to reduce the capacity of the crop to produce seed yield and can be represented as a decrease in the Effective Leaf Area Index (ELAI).

The second step was to develop a yield loss model. Any model must include the effect of time of SBR arrival on final yield. SBR starting at flowering results in greater damage than if it arrives during seed fill. This is achieved by integrating ELAI over time to get Effective Leaf Area Duration (ELAD). There was a robust linear relationship between the relative change in ELAD and reduction in yield from SBR. The regression equation thus developed is the basis of the yield loss model.

The third step was to use independent data sets originating from the United States (FL and GA) and Brazil to evaluate the “reliability” of the model. These trials used a range of cultivar maturities (MG 5 to 8) and row widths (15” to 36”). The independent data sets generated affirmed the ELAD/yield relationship across a number of regions and production practices. However, these only represent a small sampling of the genotypes, row widths, growing conditions and management operations likely in the southern U.S.

The fourth step was to use the model to generate yield loss estimates and to develop a user-friendly interactive website using these estimates. Curves were fit to ELAI data from Brazil which plotted the progressive SBR damage from “severe” epidemics starting at different times. The curves were integrated over time to calculate ELAD which was then used to estimate the reduction in yields. To generate estimates for ‘moderate’ and ‘light’ epidemics, hypothetical scenarios, based on what is ‘likely’ to occur to the lag phase and the rate of disease progress under the three scenarios, were used. The yield loss estimates must be taken with the understanding of these limitations.

More detailed information about the development of the risk management tool is available in this PDF document.

An Extension Plant Pathology Fact Sheet titled "Soybean Yield Loss Prediction Tool for Managing Soybean Rust" which summarizes information about tool development, limitations of the model and how to use the tool is available here.

Refereed Publications:

Kumudini, S., C.V. Godoy, J.E. Board, J. Omielan, and M. Tollenaar. 2008a. Mechanisms involved in soybean rust-induced yield reduction. Crop Sci. 48:2334-2342. Link to abstract <http://crop.scijournals.org/cgi/content/abstract/48/6/2334>

Kumudini, S., E. Prior, J. Omielan, and M. Tollenaar. 2008b. Impact of Phakopsora pachyrhizi infection on soybean leaf photosynthesis and radiation absorption. Crop Sci. 48:2343-2350. Link to abstract <http://crop.scijournals.org/cgi/content/abstract/48/6/2343>