Remote sensing has been increasingly used for precision nitrogen management to assess the plant nitrogen status in a spatial and real-time manner. The nitrogen nutrition index (NNI) can quantitatively describe the nitrogen status of crops. Nevertheless, the NNI diagnosis for cotton with unmanned aerial vehicle (UAV) multispectral images has not been evaluated yet. This study aimed to evaluate the performance of three machine learning models, i.e., support vector machine (SVM), back propagation neural network (BPNN), and extreme gradient boosting (XGB) for predicting canopy nitrogen weight and NNI of cotton over the whole growing season from UAV images. The results indicated that the models performed better when the top 15 vegetation indices were used as input variables based on their correlation ranking with nitrogen weight and NNI. The XGB model performed the best among the three models in predicting nitrogen weight. The prediction accuracy of nitrogen weight at the upper half-leaf level (R²=0.89, RMSE=0.68 g m^–2, RE=14.62% for calibration and R2=0.83, RMSE=1.08 g m^–2, RE=19.71% for validation) was much better than that at the all-leaf level (R²=0.73, RMSE=2.20 g m^–2, RE=26.70% for calibration and R²=0.70, RMSE=2.48 g m^–2, RE=31.49% for validation) and at the plant level (R²=0.66, RMSE=4.46 g m^–2, RE=30.96% for calibration and R2=0.63, RMSE=3.69 g m^–2, RE=24.81% for validation). Similarly, the XGB model (R²=0.65, RMSE=0.09, RE=8.59% for calibration and R²=0.63, RMSE=0.09, RE=8.87% for validation) also outperformed the SVM model (R²=0.62, RMSE=0.10, RE=7.92% for calibration and R²=0.60, RMSE=0.09, RE=8.03% for validation) and BPNN model (R2=0.64, RMSE=0.09, RE=9.24% for calibration and R²=0.62, RMSE=0.09, RE=8.38% for validation) in predicting NNI. The NNI predictive map generated from the optimal XGB model can intuitively diagnose the spatial distribution and dynamics of nitrogen nutrition in cotton fields, which can help farmers implement precise cotton nitrogen management in a timely and accurate manner

The accurate and rapid estimation of canopy nitrogen content (CNC) in crops is the key to optimizing in-season nitrogen fertilizer application in precision agriculture. However, the determination of CNC from field sampling data for leaf area index (LAI), canopy photosynthetic pigments (CPP; including chlorophyll a, chlorophyll b and carotenoids) and leaf nitrogen concentration (LNC) can be time-consuming and costly. Here we evaluated the use of high-precision unmanned aerial vehicle (UAV) multispectral imagery for estimating the LAI, CPP and CNC of winter wheat over the whole growth period. A total of 23 spectral features (SFs; five original spectrum bands, 17 vegetation indices and the gray scale of the RGB image) and eight texture features (TFs; contrast, entropy, variance, mean, homogeneity, dissimilarity, second moment, and correlation) were selected as inputs for the models. Six machine learning methods, i.e., multiple stepwise regression (MSR), support vector regression (SVR), gradient boosting decision tree (GBDT), Gaussian process regression (GPR), back propagation neural network (BPNN) and radial basis function neural network (RBFNN), were compared for the retrieval of winter wheat LAI, CPP and CNC values, and a double-layer model was proposed for estimating CNC based on LAI and CPP. The results showed that the inversion of winter wheat LAI, CPP and CNC by the combination of SFs+TFs greatly improved the estimation accuracy compared with that by using only the SFs. The RBFNN and BPNN models outperformed the other machine learning models in estimating winter wheat LAI, CPP and CNC. The proposed double-layer models (R²=0.67–0.89, RMSE=13.63–23.71 mg g^–1, MAE=10.75–17.59 mg g^–1) performed better than the direct inversion models (R²=0.61– 0.80, RMSE=18.01–25.12 mg g^–1, MAE=12.96–18.88 mg g^–1) in estimating winter wheat CNC. The best winter wheat CNC accuracy was obtained by the double-layer RBFNN model with SFs+TFs as inputs (R²=0.89, RMSE=13.63 mg g^–1, MAE=10.75 mg g^–1). The results of this study can provide guidance for the accurate and rapid determination of winter wheat canopy nitrogen content in the field.

Maize (Zea mays L.) is one of the three major food crops and an important source of carbohydrates for maintaining food security around the world.  Plant height (H), stem diameter (SD), leaf area index (LAI) and dry matter (DM) are important growth parameters that influence maize production.  However, the combined effect of temperature and light on maize growth is rarely considered in crop growth models.  Ten maize growth models based on the modified logistic growth equation (Mlog) and the Mitscherlich growth equation (Mit) were proposed to simulate the H, SD, LAI and DM of maize under different mulching practices based on experimental data from 2015–2018.  Either the accumulative growing degree-days (AGDD), helio thermal units (HTU), photothermal units (PTU) or photoperiod thermal units (PPTU, first proposed here) was used as a single driving factor in the models; or AGDD was combined with either accumulative actual solar hours (ASS), accumulative photoperiod response (APR, first proposed here) or accumulative maximum possible sunshine hours (ADL) as the dual driving factors in the models.  The model performances were evaluated using seven statistical indicators and a global performance index.  The results showed that the three mulching practices significantly increased the maize growth rates and the maximum values of the growth curves compared with non-mulching.  Among the four single factor-driven models, the overall performance of the Mlog_PTU Model was the best, followed by the Mlog_AGDD Model.  The Mlog_PPTU Model was better than the Mlog_AGDD Model in simulating SD and LAI.  Among the 10 models, the overall performance of the Mlog_AGDD–APR Model was the best, followed by the Mlog_AGDD–ASS Model.  Specifically, the Mlog_AGDD–APR Model performed the best in simulating H and LAI, while the Mlog_AGDD–ADL and Mlog_AGDD–ASS models performed the best in simulating SD and DM, respectively.  In conclusion, the modified logistic growth equations with AGDD and either APR, ASS or ADL as the dual driving factors outperformed the commonly used modified logistic growth model with AGDD as a single driving factor in simulating maize growth.