No abstract

We used the Agricultural Production Systems sIMulator (APSIM) to predict and explain maize and soybean yields, phenology, and soil water and nitrogen (N) dynamics during the growing season in Iowa, USA. Historical, current and forecasted weather data were used to drive simulations, which were released in public four weeks after planting. In this paper, we (1) describe the methodology used to perform forecasts;(2) evaluate model prediction accuracy against data collected from 10 locations over four years; and (3) identify inputs that are key in forecasting yields and soil N dynamics. We found that the predicted median yield at planting was a very good indicator of end-of-season yields (relative root mean square error [RRMSE] of ∼20%). For reference, the prediction at maturity, when all the weather was known, had a RRMSE of 14%. The good prediction at planting time was explained by the existence of shallow water tables, which decreased model sensitivity to unknown summer precipitation by 50-64%. Model initial conditions and management information accounted for Abbreviations: APSIM, Agricultural Production Systems sIMulator; RRMSE, relative root mean square error.This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes. one-fourth of the variation in maize yield. End of season model evaluations indicated that the model simulated well crop phenology (R 2 = 0.88), root depth (R 2 = 0.83), biomass production (R 2 = 0.93), grain yield (R 2 = 0.90), plant N uptake (R 2 = 0.87), soil moisture (R 2 = 0.42), soil temperature (R 2 = 0.93), soil nitrate (R 2 = 0.77), and water table depth (R 2 = 0.41). We concluded that model set-up by the user (e.g. inclusion of water table), initial conditions, and early season measurements are very important for accurate predictions of soil water, N and crop yields in this environment. Neil Huth from CSIRO for their support with the APSIM model, Iowa State University students () for assistance with data collection and managing the field experiments. We also thank the APSIM Initiative for making the software publicly available and for ensuring software quality. ORCIDSotirios V. Archontoulis https://orcid.org/0000-0001-7595-8107 Mark A. Licht https://orcid.org/0000-0001-6640-7856 Kendall R. Lamkey

No abstract

Pre-growing season prediction of crop production outcomes such as grain yields and nitrogen (N) losses can provide insights to farmers and agronomists to make decisions. Simulation crop models can assist in scenario planning, but their use is limited because of data requirements and long runtimes. Thus, there is a need for more computationally expedient approaches to scale up predictions. We evaluated the potential of four machine learning (ML) algorithms (LASSO Regression, Ridge Regression, random forests, Extreme Gradient Boosting, and their ensembles) as meta-models for a cropping systems simulator (APSIM) to inform future decision support tool development. We asked:(1) How well do ML meta-models predict maize yield and N losses using pre-season information? (2) How many data are needed to train ML algorithms to achieve acceptable predictions? (3) Which input data variables are most important for accurate prediction? And (4) do ensembles of ML meta-models improve prediction? The simulated dataset included more than three million data including genotype, environment and management scenarios. XGBoost was the most accurate ML model in predicting yields with a relative mean square error (RRMSE) of 13.5%, and Random forests most accurately predicted N loss at planting time, with a RRMSE of 54%. ML meta-models reasonably reproduced simulated maize yields using the information available at planting, but not N loss. They also differed in their sensitivities to the size of the training dataset. Across all ML models, yield prediction error decreased by 10%-40% as the training dataset increased from 0.5 to 1.8 million data points, whereas N loss prediction error showed no consistent pattern. ML models also differed in their sensitivities to input variables (weather, soil properties, management, initial conditions), thus depending on the data availability researchers may use a different ML model. Modest prediction improvements resulted from ML ensembles. These results can help accelerate progress in coupling simulation models and ML toward developing dynamic decision support tools for pre-season management.

In addition to soil health and conservation benefits, cover crops (CCs) may offer weed control in the midwestern United States, but individual studies report varying effects. We conducted a meta-analysis of studies measuring weed biomass (WBIO) or density (WDEN) in paired CC and no-cover treatments in corn (Zea mays L.)-soybean [Glycine max (L.) Merr] rotations in the U.S. Midwest. Fifteen studies provided 123 paired comparisons of WBIO and 119 of WDEN. Only grass CCs significantly reduced WBIO, while no CC reduced WDEN. We found no evidence CC management factors (e.g., termination method) directly affected outcomes. Our dataset showed that a 75% reduction in WBIO requires at least 5 Mg ha −1 of CC. Simulations from a process-based model (SALUS) indicated achieving 5 Mg ha −1 requires substantially earlier fall planting and later spring termination in most years, conflicting with typical cash-crop planting and harvesting. We conclude CCs significantly reduce WBIO, but current CC management constraints render these reductions variable and uncertain.

No abstract

Quantitative measurements of root traits can improve our understanding of how crops respond to soil and weather conditions, but such data are rare. Our objective was to quantify maximum root depth and root front velocity (RFV) for maize (Zea mays) and soybean (Glycine max) crops across a range of growing conditions in the Midwest USA. Two sets of root measurements were taken every 10-15 days: in the crop row (in-row) and between two crop rows (center-row) across six Iowa sites having different management practices such as planting dates and drainage systems, totaling 20 replicated experimental treatments. Temporal root data were best described by linear segmental functions. Maize RFV was 0.62 ± 0.2 cm d −1 until the 5th leaf stage when it increased to 3.12 ± 0.03 cm d −1 until maximum depth occurred at the 18th leaf stage (860°Cd after planting). Similar to maize, soybean RFV was 1.19 ± 0.4 cm d −1 until the 3rd node when it increased to 3.31 ± 0.5 cm d −1 until maximum root depth occurred at the 13th node (813.6°C d after planting). The maximum root depth was similar between crops (P > 0.05) and ranged from 120 to 157 cm across 18 experimental treatments, and 89-90 cm in two experimental treatments. Root depth did not exceed the average water table (two weeks prior to start grain filling) and there was a significant relationship between maximum root depth and water table depth (R 2 = 0.61; P = 0.001). Current models of root dynamics rely on temperature as the main control on root growth; our results provide strong support for this relationship (R 2 > 0.76; P < 0.001), but suggest that water table depth should also be considered, particularly in conditions such as the Midwest USA where excess water routinely limits crop production. These results can assist crop model calibration and improvements as well as agronomic assessments and plant breeding efforts in this region.

Root traits are important to crop functioning, yet there is little information about how root traits vary with shoot traits. Using a standardized protocol, we collected 160 soil cores (0−210 cm) across 10 locations, three years and multiple cropping systems (crops x management practices) in Iowa, USA. Maximum root biomass ranged from 1.2 to 2.8 Mg ha−1 in maize and 0.86 to 1.93 Mg ha−1 in soybean. The root:shoot (R:S) ratio ranged from 0.04 to 0.13 in maize and 0.09 to 0.26 in soybean. Maize produced 27 % more root biomass, 20 % longer roots, with 35 % higher carbon to nitrogen (C:N) ratio than soybean. In contrast, soybean had a 47 % greater R:S ratio than maize. The maize R:S ratio values were substantially lower than literature values, possibly due to differences in measurement methodologies, genotypes, and environment. In particular, we sampled at plant maturity rather than crop harvest to minimize the effect of senescence on measurements of shoots and roots. Maximum shoot biomass explained 70 % of the variation in root biomass, and the R:S ratio was positively correlated with the root C:N measured in both crops. Easily-measured environmental variables including temperature and precipitation were weakly associated with root traits. These results begin to fill an important knowledge gap that will enable better estimates of belowground net primary productivity and soil organic matter dynamics. Ultimately, the ability to explain variation in root mass production can be used to improve C and N budgets and modeling studies from crop to regional scales.