Yield gaps of maize (Zea mays L.) in the smallholder farms of eastern India are outcomes of a complex interplay of climatic variations, soil fertility gradients, socio-economic factors, and differential management intensities. Several machine learning approaches were used in this study to investigate the relative influences of multiple biophysical, socio-economic, and crop management features in determining maize yield variability using several machine learning approaches. Soil fertility status was assessed in 180 farms and paired with the surveyed data on maize yield, socio-economic conditions, and agronomic management. The C&RT relative variable importance plot identified farm size, total labor, soil factors, seed rate, fertilizer, and organic manure as influential factors. Among the three approaches compared for classifying maize yield, the artificial neural network (ANN) yielded the least (25%) misclassification on validation samples. The random forest partial dependence plots revealed a positive association between farm size and maize productivity. Nonlinear support vector machine boundary analysis for the eight top important variables revealed complex interactions underpinning maize yield response. Notably, farm size and total labor synergistically increased maize yield. Future research integrating these algorithms with empirical crop growth models and crop simulation models for ex-ante yield estimations could result in further improvement.

It is critical to understand nutrient dynamics within different plant parts to correctly fine-tune agronomic advices, and to update breeding programs for increasing nutrient use efficiencies and yields. Farmer’s field-based research was conducted to assess the effects of nitrogen (N), phosphorus (P), and potassium (K) levels on dry matter and nutrient accumulation, partitioning, and remobilization dynamics in three popular maize (Zea mays L.) hybrids (P3522, P3396, and Rajkumar) over two years in an alluvial soil of West Bengal, India. Experimental results revealed that NPK rates as well as different cultivars significantly (p ≤ 0.05) influenced the dry matter accumulation (DMA) in different plant parts of maize at both silking and physiological maturity. The post-silking dry matter accumulation (PSDMA) and post-silking N, P, and K accumulations (PSNA, PSPA, PSKA) were highest in cultivar P3396. However, cultivar P3522 recorded the highest nutrient remobilizations and contributions to grain nutrient content. Total P and K accumulation were highest with 125% of the recommended dose of fertilizer (RDF) while total N accumulation increased even after 150% RDF (100% RDF is 200 kg N, 60 kg P₂O₅, and 60 kg K₂O ha^–1 for the study region). Application of 125% RDF was optimum for PSDMA. The PSNA continued to increase up to 150% RDF while 125% RDF was optimum for PSPA. Cultivar differences significantly affected both remobilization efficiency (RE) and contribution to grain nutrient content for all tested macronutrients (N, P, and K). In general, RE as well as contribution to grain nutrient content was highest at 125% RDF for N and K, and at 100% RDF for P (either significantly or at par with other rates) for plots receiving nutrients. For all tested cultivars, nutrient remobilization and contribution to grain nutrient content was highest under nutrient-omission plots and absolute control plots. Both year and cultivar effects were non-significant for both grain and stover yields of maize. Application of 75% RDF was sufficient to achieve the attainable yield at the study location. The cultivar P3522 showed higher yield over both P3396 and Rajkumar, irrespective of fertilizer doses, although, the differences were not statistically significant (p ≥ 0.05). The study underscores the importance of maize adaptive responses in terms of nutrients accumulation and remobilization at different levels of nutrient availability for stabilizing yield.

In the present two-year study, an attempt was made to estimate the grain yield, grain nutrient uptake, and oil quality of three commonly grown maize (Zea mays L.) hybrids fertilized with varied levels of nitrogen (N), phosphorus (P) and potassium (K). Results obtained from both the experimental years indicated that application of 125% of recommended dose of fertilizer (RDF) recorded maximum grain yield (10.37 t ha^-1; 124% higher than control). When compared with 100% RDF, grain yield reduction with nutrient omission was 44% for N omission, 17% for P omission, and 27% for K omission. Nitrogen uptake was increased with increasing NPK levels up to 150% RDF that was statistically at par (p ≥ 0.01) with 125% RDF. Increasing trend in P and K uptake was observed with successive increase in NPK levels up to 125% RDF, above which it declined. The protein content was significantly higher in grains of var. P 3396 with 125% RDF. Nutrient management has significant (p ≤ 0.01) role in the grain oil content. Saturated fatty acids (palmitic, stearic and arachidic acid) content decreased, and unsaturated fatty acid (oleic, linoleic and linolenic acid) increased with increasing NPK levels. The average oleic acid desaturation and linoleic acid desaturation ratios were increased with increasing NPK levels up to 100 and 125% RDF, respectively. However, average monounsaturated fatty acids (MUFA): poly-unsaturated fatty acids (PUFA), saturated: unsaturated as well as linoleic: linolenic acid ratios were increased on receiving 75% RDF, and beyond that it showed decreasing trend. The omission of K had the highest inhibitory effect on corn oil quality followed by N and P omission.

Rice is the lifeline for more than half of the world population, and in India, in view of its huge demand in the country, farmers adopt a rice–rice cropping system where the irrigation facility is available. As rice is a nutrient-exhausting crop, sustainable productivity of rice–rice cropping system greatly depends on appropriate nutrient management in accordance with the inherent soil fertility. The application of an ample dose of fertilizer is the key factor for maintaining sustainable rice yields and nutrient balance of the soil. Considering the above facts, an experiment was conducted on nutrient management in a rice–rice cropping system at the university farm of Visva-Bharati, situated in a sub-tropical climate under the red and lateritic belt of the western part of West Bengal, India, during two consecutive years (2014–2016). The experiment was laid out in a Randomized Completely Block Design with 12 treatments and three replications, with different rates of N:P:K:Zn:S application in both of the growing seasons, namely, kharif and Boro. The recommended (ample) dose of nutrients was 80:40:40:25:20 and 120:60:60:25:20 kg ha⁻¹ of N:P₂O₅:K₂O:Zn:S in the Kharif and Boro season, respectively. A high yielding variety, named MTU 7029, and a hybrid, Arize 6444 GOLD, were taken in the Kharif and Boro seasons, respectively. The results clearly indicated that the application of a recommended dose of nutrients showed its superiority over the control (no fertilizer application) in the expression of growth characters, yield attributes, yields, and nutrient uptake of Kharif as well as Boro rice. Out of the all treatments, the best result was found in the treatment where the ample dose of nutrients was applied, resulting in maximum grain yield in both the Kharif (5.6 t ha⁻¹) and Boro (6.6 t ha⁻¹) season. The corresponding yield attributes for the same treatment in the Kharif (panicles m⁻²: 247.9; grains panicle⁻¹: 132.0; spikelets panicle⁻¹: 149.6; test weight: 23.8 g; and panicle length: 30.6 cm) and Boro (panicles m⁻²: 281.6; grains panicle⁻¹: 142.7; spikelets panicle⁻¹: 157.2; test weight: 24.8 g; and panicle length: 32.8 cm) season explained the maximum yield in this treatment. Further, a reduction or omission of individual nutrients adversely impacted on the above traits and resulted in a negative balance of the respective nutrients. The study concluded that the application of a recommended dose of nutrients was essential for proper nutrient balance and sustainable yields in the rice–rice cropping system.