Machine Learning in Evaluating Multispectral Active Canopy Sensor for Prediction of Corn Leaf Nitrogen Concentration and Yield

Applying the optimum rate of fertilizer nitrogen (N) is a critical factor for field management. Multispectral information collected by active canopy sensors can potentially indicate the leaf N status and aid in predicting grain yield. Crop Circle multispectral data were acquired with the purpose of measuring the reflectance data to calculate vegetation indices (VIs) at different growth stages. Applying the optimum rate of fertilizer N can have a considerable impact on grain yield and profitability. The objectives of this study were to evaluate the reliability of a handheld Crop Circle ACS-430, to estimate corn leaf N concentration and predict grain yield of corn using machine learning (ML) models. The analysis was conducted using four ML models to identify the best prediction model for measurements acquired with a Crop Circle ACS-430 field sensor at three growth stages. Four fertilizer N levels from deficient to excessive in 50/50 spilt were applied to corn at 1–2 leaves, with visible leaf collars (V1-V2 stage) and at the V6-V7 stage to establish widely varying N nutritional status. Crop Circle spectral observations were used to derive 25 VIs for different growth stages (V4, V6, and VT) of corn at the W. B. Andrews Agricultural Systems farm of Mississippi State University. Multispectral raw data, along with Vis, were used to quantify leaf N status and predict the yield of corn. In addition, the accuracy of wavelength-based and VI-based models were compared to examine the best model inputs. Due to limited observed data, the stratification approach was used to split data to train and test set to obtain balanced data for each stage. Repeated cross validation (RCV) was then used to train the models. Results showed that the Simplified Canopy Chlorophyll Content Index (SCCCI) and Red-edge ratio vegetation index (RERVI) were the most effective VIs for estimating leaf N% and that SCCCI, Red-edge chlorophyll index (CIRE), RERVI, Soil Adjusted Vegetation Index (SAVI), and Normalized Difference Vegetation Index (NDVI) were the most effective VIs for predicting corn grain yield. Additionally, among the four ML models utilized in this research, support vector regression (SVR) achieved the most accurate results for estimating leaf N concentration using either spectral bands or VIs as the model inputs.

Solar radiation effects on leaf nitrogen and phosphorus stoichiometry of Chinese fir across subtropical China

Background Solar radiation (SR) plays critical roles in plant physiological processes and ecosystems functions. However, the exploration of SR influences on the biogeochemical cycles of forest ecosystems is still in a slow progress, and has important implications for the understanding of plant adaption strategy under future environmental changes. Herein, this research was aimed to explore the influences of SR on plant nutrient characteristics, and provided theoretical basis for introducing SR into the establishment of biochemical models of forest ecosystems in the future researches. Methods We measured leaf nitrogen (N) and phosphorus (P) stoichiometry in 19 Chinese fir plantations across subtropical China by a field investigation. The direct and indirect effects of SR, including global radiation (Global R), direct radiation (Direct R) and diffuse radiation (Diffuse R) on the leaf N and P stoichiometry were investigated. Results The linear regression analysis showed that leaf N concentration had no association with SR, while leaf P concentration and N:P ratio were negatively and positively related to SR, respectively. Partial least squares path model (PLS-PM) demonstrated that SR (e.g. Direct R and Diffuse R), as a latent variable, exhibited direct correlations with leaf N and P stoichiometry as well as the indirect correlation mediated by soil P content. The direct associations (path coefficient = − 0.518) were markedly greater than indirect associations (path coefficient = − 0.087). The covariance-based structural equation modeling (CB-SEM) indicated that SR had direct effects on leaf P concentration (path coefficient = − 0.481), and weak effects on leaf N concentration. The high SR level elevated two temperature indexes (mean annual temperature, MAT; ≥ 10 °C annual accumulated temperature, ≥ 10 °C AAT) and one hydrological index (mean annual evapotranspiration, MAE), but lowered the soil P content. MAT, MAE and soil P content could affect the leaf P concentration, which cause the indirect effect of SR on leaf P concentration (path coefficient = 0.004). Soil N content had positive effect on the leaf N concentration, which was positively and negatively regulated by MAP and ≥ 10 °C AAT, respectively. Conclusions These results confirmed that SR had negatively direct and indirect impacts on plant nutrient status of Chinese fir based on a regional investigation, and the direct associations were greater than the indirect associations. Such findings shed light on the guideline of taking SR into account for the establishment of global biogeochemical models of forest ecosystems in the future studies.

Down-Regulation of Photosynthesis to Elevated CO2 and N Fertilization in Understory Fraxinus rhynchophylla Seedlings

(1) Background: Down-regulation of photosynthesis has been commonly reported in elevated CO2 (eCO2) experiments and is accompanied by a reduction of leaf nitrogen (N) concentration. Decreased N concentrations in plant tissues under eCO2 can be attributed to an increase in nonstructural carbohydrate (NSC) and are possibly related to N availability. (2) Methods: To examine whether the reduction of leaf N concentration under eCO2 is related to N availability, we investigated understory Fraxinus rhynchophylla seedlings grown under three different CO2 conditions (ambient, 400 ppm [aCO2]; ambient × 1.4, 560 ppm [eCO21.4]; and ambient × 1.8, 720 ppm [eCO21.8]) and three different N concentrations for 2 years. (3) Results: Leaf and stem biomass did not change under eCO2 conditions, whereas leaf production and stem and branch biomass were increased by N fertilization. Unlike biomass, the light-saturated photosynthetic rate and photosynthetic N-use efficiency (PNUE) increased under eCO2 conditions. However, leaf N, Rubisco, and chlorophyll decreased under eCO2 conditions in both N-fertilized and unfertilized treatments. Contrary to the previous studies, leaf NSC decreased under eCO2 conditions. Unlike leaf N concentration, N concentration of the stem under eCO2 conditions was higher than that under ambient CO2 (4). Conclusions: Leaf N concentration was not reduced by NSC under eCO2 conditions in the understory, and unlike other organs, leaf N concentration might be reduced due to increased PNUE.

Feasibility Study of Using Absolute SPAD Values for Standardized Evaluation of Corn Nitrogen Status

Nitrogen fertilizer recommendations for corn (Zea mays L.) should ensure high yields using adequate N doses. Soil–plant analysis development (SPAD) meter technology using absolute SPAD values, might be more reliable than relative SPAD values in helping corn producers making timely decisions about N applications. This study aimed to develop a relationship between absolute SPAD values and leaf N concentration, and to determine optimal leaf N concentrations at early corn growth stages (V6, V8, V10, and V12). Three experiments were conducted during two summer seasons (2014 and 2015) using six N treatments applied at early corn growth stages. In parallel, two experiments were carried out in a high residual N environment to establish the optimum corn leaf N concentration. Results showed a significant linear relationship between corn leaf N concentrations and absolute SPAD values (R2 = 0.80, p < 0.05). The mean optimum corn leaf N concentration decreased over corn growth stages. It is of great importance to make the absolute SPAD method accessible for farmers, but more research is required to perform standardized reading of absolute SPAD values data.

Slow Release Fertilizer has no Effect on Soil and Plant Nitrogen and Fruit Yield in Bell Pepper (Capsicum annum L.)

Aims: To evaluate the effects of combining slow release fertilizer (SRF) with conventional N fertilizer on the levels of soil nitrogen (N), leaf N, and yield of bell pepper plants. Study Design: The design was a randomized complete block with a factorial arrangement. There were four treatments [2 N fertilizers x 2 N rates (200 kg/ha N and 280 kg/ha N)] and four replications. The N fertilizers treatments were calcium nitrate + SRF and calcium nitrate (CN) alone as the control. The rate treatments were 200 and 280 kg/ha N. Place and Duration of Study: Horticulture Farm, Department of Horticulture, Tifton Campus, University of Georgia, spring of 2008. Methodology: Bell pepper ‘Heritage’ (Harris Moran, Modesto, Calif., U.S.) transplants were planted on 10 Apr. 2008 in two rows of plants per bed, with a distance between plants of 0.30 m. Soil and leaf nitrogen and other nutrients were monitored during the season. Results: There were no consistent differences in the concentrations of NH4+-N and NO3--N at both 0-30 cm and 30-60 cm soil depth between CN + SRF and CN alone. Leaf N concentrations 40 DAT and 68 DAT were higher in plants fertilized with CN + SRF compared to the control, while there were no differences in leaf N concentration 98 DAT. Leaf N concentration was increased at the highest N fertilization rate. Marketable and total fruit yields and individual fruit weight were unaffected by fertilizer treatment and N rate. Conclusion: Utilization of a slow-release fertilizer (combined with calcium nitrate) had no benefit in reducing soil N losses or in increasing leaf N status and bell pepper fruit yields.

Nitrogen Management Based on Visible/Near Infrared Spectroscopy in Pear Orchards

The timely estimation of nitrogen (N) requirements is essential for managing N fertilizer application in pear orchards. Visible/near infrared spectroscopy is a non-destructive and effective technique for real-time assessment of leaf N concentration, but its utility for decisions about fertilizer application in the pear orchards remains to be determined. In this study, we used leaf spectroscopy to determine leaf N concentration, used this value to calculate the amounts of N required for supplementary fertilization, and then evaluated the effects of the application. Over the two-year study, Cuiguan pear trees were treated with N at the following rates: 0 (N0), 100 (N1), 200 (N2), 300 (N3), and 400 (N4) g N per tree, regarded as five “controlled” N application rates. Another four “regulatory” treatments (Nr1-4) were fertilized as the “controlled” N application rates the first year, then given adjusted N application by topdressing as calculated using the N concentrations inferred from visible/near infrared spectroscopy data the second year. A model (R2 = 0.82) was established the first year to relate leaf spectra and N concentration using a partial least squares regression with full bands (350–2500 nm). The amount of N in the topdressing for the supplemental treatments was determined using the predicted leaf N concentration and the topdressing calculation method adapted from the N balance formula. Results showed that adjusted N applications of the Nr1 and Nr2 increased yield by 26% and 23%, respectively, over the controlled treatments N1 and N2. Although treatments Nr3 and Nr4 did not increase yield significantly over N3 and N4, the partial factor productivity of nitrogen in Nr4 was higher than the N4. The transverse diameter of fruit from Nr1 trees was significantly higher than from N1 trees, while the longitudinal diameter of fruit from Nr1, Nr2, and Nr3 trees was significantly higher than from N1, N2 and N3 trees, suggesting that fruit longitudinal growth and single-fruit weight is stimulated by adjusted N applications. However, the soluble solids in fruit from trees receiving adjusted N were not significantly greater than in fruit from non-supplemented trees. In conclusion, our results illustrate that regulatory N management contributes to fruit yield and quality especially in the nitrogen deficiency condition and improves the nitrogen use efficiency in nitrogen surplus. The N prediction model established using the nondestructive visible/near infrared spectroscopy is convenient and economical.

Individual and Combined Use of Sawdust and Weed Mat Mulch in a New Planting of Northern Highbush Blueberry. II. Nutrient Uptake and Allocation

A 2-year trial was established in Oct. 2016 in western Oregon to evaluate the effects of various in-row mulch treatments on establishment of northern highbush blueberry (Vaccinium corymbosum L. ‘Duke’). The treatments included douglas fir [Pseudotsuga menziesii (Mirb.) Franco] sawdust, black weed mat (woven polypropylene groundcover), green weed mat, and sawdust covered with black or green weed mat. For the most part, plant nutrient concentration and content were unaffected by the color of the weed mat. In both years, mulching with weed mat over sawdust reduced soil NO3-N compared with weed mat alone. The only other soil nutrient affected by mulch was K, which was highest with sawdust mulch and intermediate with black weed mat alone in year 2. There were inconsistent effects of mulch on leaf nutrient concentration during the study. In 2018, leaf N concentration was lowest with black weed mat over sawdust. There were few mulch effects on nutrient concentrations in senescent leaves in both years and in harvested fruit in year 2. Mulch had greater effect on nutrient concentration in dormant plant parts after the second growing season than after the first, with the addition of sawdust under weed mat leading to significant differences for many nutrients in various plant parts compared with weed mat alone. Total uptake of N ranged from 12 kg·ha−1 (black weed mat) to 17 kg·ha−1 (black weed mat over sawdust) in year 1 and averaged 33 kg·ha−1 in year 2, with no effect of mulch. Fertilizer use efficiency for N was 8% to 12% in year 1 and 42% in year 2. Uptake of other nutrients was unaffected by mulch and, depending on the year, ranged from 1.3 to 4.3 kg·ha−1 P, 4.0 to 8.0 kg·ha−1 K, 2.1 to 4.9 kg·ha−1 Ca, and 1.0 to 1.5 kg·ha−1 Mg. Each of these other nutrients was derived from the soil or decomposing roots.

Variation in Leaf Nitrogen and Phosphorus Stoichiometry in Different Functional Groups of Legumes

The legume is notable owing to their symbiotic nitrogen (N) fixing ability. Usually, higher leaf N concentration and N to phosphorus (P) ratio (N:P) in legumes than non-legumes. However, the variations of leaf N, P and N:P and their relationship had been hardly studied based on functional groups. In this study, we studied the leaf N, P and N:P and their relationship among different functional groups. The results showed that the average values of leaf N, P and N:P ratios for all legumes were 27.33 mg g-1, 1.27 mg g-1 and 21.94, respectively. Leaf N (36.96 mg g-1) and P (2.15 mg g-1) of herbaceous legumes are significantly higher than N (24.97 mg g-1) and P (1.18 mg g-1) in woody plants, respectively. Moreover, leaf N, P and N:P of shrub markedly higher than them in tree. Leaf N and P are always higher in deciduous than evergreen legumes. A negative correlation was found between leaf N:P and P in overall and different functional groups of legumes.

Estimating Plant Nitrogen Concentration of Maize Using a Leaf Fluorescence Sensor across Growth Stages

Nitrogen (N) is one of the most essential nutrients that can significantly affect crop grain yield and quality. The implementation of proximal and remote sensing technologies in precision agriculture has provided new opportunities for non-destructive and real-time diagnosis of crop N status and precision N management. Notably, leaf fluorescence sensors have shown high potential in the accurate estimation of plant N status. However, most studies using leaf fluorescence sensors have mainly focused on the estimation of leaf N concentration (LNC) rather than plant N concentration (PNC). The objectives of this study were to (1) determine the relationship of maize (Zea mays L.) LNC and PNC, (2) evaluate the main factors influencing the variations of leaf fluorescence sensor parameters, and (3) establish a general model to estimate PNC directly across growth stages. A leaf fluorescence sensor, Dualex 4, was used to test maize leaves with three different positions across four growth stages in two fields with different soil types, planting densities, and N application rates in Northeast China in 2016 and 2017. The results indicated that the total leaf N concentration (TLNC) and PNC had a strong correlation (R2 = 0.91 to 0.98) with the single leaf N concentration (SLNC). The TLNC and PNC were affected by maize growth stage and N application rate but not the soil type. When used in combination with the days after sowing (DAS) parameter, modified Dualex 4 indices showed strong relationships with TLNC and PNC across growth stages. Both modified chlorophyll concentration (mChl) and modified N balance index (mNBI) were reliable predictors of PNC. Good results could be achieved by using information obtained only from the newly fully expanded leaves before the tasseling stage (VT) and the leaves above panicle at the VT stage to estimate PNC. It is concluded that when used together with DAS, the leaf fluorescence sensor (Dualex 4) can be used to reliably estimate maize PNC across growth stages.