Assessing performance of empirical models for forecasting crop responses to variable fertilizer rates using on-farm precision experimentation

Abstract

Data-driven decision making in agriculture can be augmented by utilizing the data gathered from precision agriculture technologies to make the most informed decisions that consider spatiotemporal specificity. Decision support systems utilize underlying models of crop responses to generate management recommendations, yet there is uncertainty in the literature on the best model forms to characterize crop responses to agricultural inputs likely due for the most part to the variability in crop responses to input rates between fields and across years. Seven fields with at least three years of on-farm experimentation, in which nitrogen fertilizer rates were varied across the fields, were used to compare the ability of five different model types to forecast crop responses and net-returns in a year unseen by the model. All five model types were fit for each field using all permutations of the three years of data where two years were used for training and a third was held out to represent a “future” year. The five models tested were a frequentist based non-linear sigmoid function, a generalized additive model, a non-linear Bayesian regression model, a Bayesian multiple linear regression model and a random forest regression model. The random forest regression typically resulted in the most accurate forecasts of crop responses and net-returns across most fields. However, in some cases the model type that produced the most accurate forecast of grain yield was not the same as the model producing the most accurate forecast of grain protein concentration. Models performed best when the data used for training models was collected from years with similar weather conditions to the forecasted year. The results are important to developers of decision support tools because the underlying models used to simulate management outcomes and calculate net-returns need to be selected with consideration for the spatiotemporal specificity of the data available.

Using spatially variable nitrogen application and crop responses to evaluate crop nitrogen use efficiency

“…A set of six potential generalizable models for estimating NUE at a subfield scale were compared to explore the potential for low-cost methods of developing fertilizer application prescriptions. Development of a general model for predicting subfield NUE relied on environmental covariates gathered from Google Earth Engine (GEE), which were assumed to influence NUE ( Information from each covariate was extracted to each soil sample location to form the analysis dataset (Hegedus et al 2022;Hegedus and Maxwell 2022b). Temporal data were gathered, delineated by year, and constrained in a range to the point in time that a farmer needs to make decisions on N fertilizer management of dryland winter-wheat in Montana (Hegedus and Maxwell 2022b).…”

On-Farm Precision Experiments (OFPE) framework: tapping local data to optimize crop sub-field scale decisions

“…The OFPE framework has also been applied in certified organic fields to identify optimized cash crop and cover crop seeding rates based on maximizing profit from wheat grain yields. Adoption of the OFPE framework by the Data Intensive Farm Management (DIFM) project's trials in eight states and multiple countries [20] demonstrates the flexibility and adaptability of the approach, as the fieldspecific nature of the methodology relies only on data from a specific field and performs model selection to identify the form that best characterizes crop responses in a specific field [21].…”

“…Second, all variables used to predict crop responses are from open-source or farmer owned data that are available up to the time of application decision [19]. Third, predictive ecological models built specifically for each field evolve as new data are collected in subsequent years capturing temporal dynamics due to weather and economic variability [21]. Fourth, the manager can simulate management outcomes given different weather and price conditions that most closely match the current year when a decision is made or explore a possible range of outcomes under different assumed conditions.…”

“…However, the number of years that experimentation is required may vary by field, depending on the uncertainty introduced by climate change induced increases in the probability of extreme weather events and shifting temporal trends in precipitation and temperature [26]. A backlog of experimental data from years with a diverse range of weather conditions increases the likelihood that a given model can accurately simulate crop responses in unknown conditions of future years [21].…”

“…Additionally, when modeling crop responses, the bias-variance tradeoff must be considered. Crop responses to varying agronomic input rates do not always present a clear pattern, meaning that assumptions of a specific functional form or shape increase uncertainty and lead to models with reduced ability to accurately predict future crop responses compared to models that do not assume data takes a specific shape [21]. Increased predictive power translates into a higher confidence in management recommendations and evaluation of optimized input rates [19,21].…”