Update:
Objective 1 – expand coverage of FACTS

Regarding the modeling part of the project, during the 2nd project year the ISU team did the followings: 1) performed real-time field scale forecasts for eight Iowa fields; 2) performed real-time regional scale predictions for the three I-states https://crops.extension.iastate.edu/facts/regional-scale; 3) developed new algorithms to improve prediction of soybean yields under excess water stress; 4) finalized the development of the Next Generation Soybean Crop Growth model in collaboration with CSRIO.

Regarding the field scale FACTS prediction (on-line) tool we initiated the predictions in June of 2019 and we updated the predictions every 2-weeks since then. Each time we released the following information to Iowa farmers: crop staging, yield predictions, soil water and nitrogen by layer, and crop water and nitrogen uptake including “benchmarks”. The accuracy of the 2019 predictions was in general very good (less than 5-10% error).

The most important accomplishment of our modeling work was the release of the regional scale forecast that covers Iowa, Illinois, and Indiana farmers. The forecast was updated weekly, every Wednesday morning, and included the following information: soil water index (2019 soil moisture relative to field capacity), soil nitrogen mineralization index (2019 soil mineralization relative to 35-yr average), soil temperature, crop growth rates, crop nitrogen uptake, crop N accumulation, crop photosynthetic rate and root depth. We covered 30 fields per county for a total of 9,000 soybean fields across Iowa, Illinois and Indiana. For the year 2019 we performed 18 forecasts (June to Oct). Fig 1 in the supplementary materials show model performance in simulating yields for selected counties within the I-states (slope = 1; RMSE = 7.1 bu/ac, RMSE = 15.4%).

With regards to model improvement, we synthesized information from 8 literature papers on flooding (duration and timing) response to yield, developed and incorporated new equations into the APSIM crop model, and wrote a new scientific publication which was accepted in Frontiers in Plant Science Journal (impact factor 4.1) (https://www.frontiersin.org/articles/10.3389/fpls.2020.00062/full?report=reader). Overall, this work improved APSIM prediction ability by 35% compared to the default model in situations with excess water stress. A second paper was also published regarding modeling soybean crops and predicting plant N dynamics by our group (https://acsess.onlinelibrary.wiley.com/doi/full/10.1002/csc2.20039). In both manuscripts, funding from ISA is highly acknowledged. Also, during the summer/fall 2019, we finalized the development of the new soybean model, which is now under review.

Objective 2 – high yielding experiments

With regards to the experimental objective, during the 2nd and final year we established three field experiments located in central, northwest and southwest Iowa. Each trial had normal and high input management treatments. The high input management included irrigation (as needed) and N-fertilization applications in addition to higher pre-planting PK fertilization rates. In central Iowa narrow rows were used for the high input management. Data collection included crop biomass during the season, soil moisture and nitrate data. The 2nd year results indicated a 6 bu/ac yield advantage of the high input system compared to the low input system in two of the three location (see Fig 2 in the appendix). The Sutherland location trial was affected by diseases and the yield difference between treatments were zero (see graph).

During the 2nd year of the project we published six papers. The first authors are graduate students in my lab, partially supported by ISA funds:

Kessler A, Archontoulis S, Licht M, 2019. Soybean Yield and Crop Stage Response to Planting Date and Cultivar Maturity in Iowa, USA. Agronomy J (accepted).

Nichols V, Ordóñez A, Wright E, Castellano M, Liebman M, Hatfield J, Helmers M, Archontoulis SV, 2019. Maize root distributions strongly associated with water tables in Iowa, USA. Plant Soil, https://doi.org/10.1007/s11104-019-04269-6

Martinez-Feria R, Licht MA, Ordonez RA, Hatfeld JL, Coulter JA, Archontoulis SV, 2019. Evaluating maize and soybean grain dry-down in the field with predictive algorithms and genotype-by-environment analysis. Nature Scientific Reports 9:7167 | https://doi.org/10.1038/s41598-019-43653-1

Cordova C, Castellano M, Dietzel R, Licht M, Togliatti K, Martinez-Feria R, Archontoulis S, 2019. Soybean nitrogen fixation dynamics in Iowa. Field Crops Research 236: 165–176.

Pasley HR, Huber I, Castellano MJ, Archontoulis SV, 2020. Modeling flood-induced stress in soybeans. Frontiers Plant Science 11:62, doi:10.3389/fpls.2020.00062.

Archontoulis SV, Castellano MJ, Licht MA, Nichols V, Baum M, Huber I, Martinez-Feria R, Puntel L, Ordóñez RA, Iqbal J, Wright EE, Dietzel RN, Helmers M, Vanloocke A, Liebman M, Hatfield JL, Herzmann D, Córdova SC, Edmonds P, Togliatti K, Kessler A, Danalatos G, Pasley H, Pederson C, Lamkey KR, 2020. Predicting Crop Yields and Soil-Plant Nitrogen Dynamics in the US Corn Belt. Crop Science, 1–18. https://doi.org/10.1002/csc2.20039

View uploaded report

1. Built capacity to simulate soybean growth and development across the entire corn belt (resolution 30 fields/county). The developed system can support scenario analyses and yield/environmental forecasting studies.
2. Improved capacity to predict soybean yields and root growth under partially flooding conditions.
3. Experiments with high vs low input management treatments showed some yield benefits in the high input system (~5 bushel/acre) but the effect was not consistent across 6 trials (2 years x 3 locations). We were expecting higher than 5 bu/ac yield benefit in the high input treatment.