Prediction of Early Season Nitrogen Uptake in Maize Using High-Resolution Aerial Hyperspectral Imagery

Tyler Nigon,Ce Yang,Fabián Fernández,Gabriel Dias Paiao,Joseph Knight,David Mulla

doi:10.3390/rs12081234

Abstract

The ability to predict spatially explicit nitrogen uptake (NUP) in maize (Zea mays L.) during the early development stages provides clear value for making in-season nitrogen fertilizer applications that can improve NUP efficiency and reduce the risk of nitrogen loss to the environment. Aerial hyperspectral imaging is an attractive agronomic research tool for its ability to capture spectral data over relatively large areas, enabling its use for predicting NUP at the field scale. The overarching goal of this work was to use supervised learning regression algorithms—Lasso, support vector regression (SVR), random forest, and partial least squares regression (PLSR)—to predict early season (i.e., V6–V14) maize NUP at three experimental sites in Minnesota using high-resolution hyperspectral imagery. In addition to the spectral features offered by hyperspectral imaging, the 10th percentile Modified Chlorophyll Absorption Ratio Index Improved (MCARI2) was made available to the learning models as an auxiliary feature to assess its ability to improve NUP prediction accuracy. The trained models demonstrated robustness by maintaining satisfactory prediction accuracy across locations, pixel sizes, development stages, and a broad range of NUP values (4.8 to 182 kg ha−1). Using the four most informative spectral features in addition to the auxiliary feature, the mean absolute error (MAE) of Lasso, SVR, and PLSR models (9.4, 9.7, and 9.5 kg ha−1, respectively) was lower than that of random forest (11.2 kg ha−1). The relative MAE for the Lasso, SVR, PLSR, and random forest models was 16.5%, 17.0%, 16.6%, and 19.6%, respectively. The inclusion of the auxiliary feature not only improved overall prediction accuracy by 1.6 kg ha−1 (14%) across all models, but it also reduced the number of input features required to reach optimal performance. The variance of predicted NUP increased as the measured NUP increased (MAE of the Lasso model increased from 4.0 to 12.1 kg ha−1 for measured NUP less than 25 kg ha−1 and greater than 100 kg ha−1, respectively). The most influential spectral features were oftentimes adjacent to each other (i.e., within approximately 6 nm), indicating the importance of both spectral precision and derivative spectra around key wavelengths for explaining NUP. Finally, several challenges and opportunities are discussed regarding the use of these results in the context of improving nitrogen fertilizer management.

Highlights

Nitrogen (N) fertilizer inputs are crucial for achieving high crop yields, but the loss of reactive N from agricultural systems leads to atmospheric, surface water, and groundwater pollution [1,2,3], diminishing environmental quality and human well-being [4,5]
Images captured from the Waseca whole field experiment that had a coarser 8 cm pixel size (Figure 2d–f) showed a consistently increasing trend in the 90th percentile MCARI2 up to the V14 development stage, whereas images captured from the Waseca small-plot experiment that had a finer 2–2.5 cm pixel size (Figure 2a–c) peaked by the V8 development stage
The current study showed nitrogen uptake (NUP) root mean squared error (RMSE) as low as 13.6 kg ha−1 across development stages (Figure 7), which was an 18% improvement compared to Xia et al [38]

Summary

Introduction

Nitrogen (N) fertilizer inputs are crucial for achieving high crop yields, but the loss of reactive N from agricultural systems leads to atmospheric, surface water, and groundwater pollution [1,2,3], diminishing environmental quality and human well-being [4,5]. Despite the potential environmental consequences to society, the pressure on producers to increase productivity oftentimes leads to N fertilizer applications in excess of crop requirement [6,7]. Without more efficient N fertilizer applications, the increasing global population and subsequent rising demand for food are expected to cause an increase in the loss of reactive N in the future [8]. A strategy to reduce the likelihood of reactive N loss is to apply part of the crop’s N requirement after emergence, delaying the application until crop demand is near its maximum. Grain yields and crop N use are not uniform across seasons [9], so this delayed application provides the opportunity to adapt N fertilizer rates in a dynamic manner based on the influence of weather and N cycle processes on early season crop growth or stress. Maize N requirement typically varies spatially [10], so this strategy can be more effective with variable rate/site-specific applications [11]

Methods

Results

Discussion

Conclusion