Search Results (38)

Ozone pollution decreases rice yield and quality in general, but how ozone stress changes grain-filling capacity is unclear. A chamber experiment was conducted to compare the effects of ozone exposure during the rice growth season on the grain-filling capacity and quality of spikelets located on the upper primary rachis (superior spikelets, SS) and the lower secondary rachis (inferior spikelets, IS). Ozone stress significantly decreased filled grain percentage by 41.4% and grain mass by 10.2% in IS, but had little effect on grain-filling capacity in SS. Consistent with the reduction in grain mass, ozone stress decreased grain volume, mainly due to reduced grain thickness, and IS was reduced more than SS. After removing the hull, brown rice obtained from ozone treatment exhibited higher proportions of immature and abnormal kernels, resulting in a substantially lower proportion of perfect kernels. Under ozone stress, the proportion of perfect kernels was only one-third in IS, compared with two-thirds in SS. Ozone stress affected the pasting properties of brown rice for both SS and IS, as shown by the decreased amylose content, and the increased maximum viscosity, minimum viscosity, final viscosity, setback, and peak time of the rapid visco analyzer profile. Out of fourteen traits related to nutritional quality of brown rice, only five showed significant increases under ozone stress, and they were the concentrations of albumin, prolamin, sulfur, copper, and manganese. The differential ozone responses between SS and IS were rather small for rice pasting properties and chemical compositions as shown by very few significant interactions between ozone and grain position. It is concluded that ozone stress during plant growth imposed more adverse effects on IS than SS in terms of grain-filling capacity and appearance quality, suggesting an enlarged asynchronous grain-filling pattern in rice panicles under ozone pollution. Strategies to improve the grain-filling capacity of IS are needed to mitigate ozone-induced damage to rice production. Full article

Rice yield and quality decline due to excessive fertiliser use is problematic in China. To increase rice grain filling and improve rice yield and quality, a nitrogen reduction and density increase study in 2023 and 2024 was imposed on a long-term experimental field. The four treatments adopted for the study were normal nitrogen and normal density (CK), normal nitrogen and increased density (NN+ID), reduced nitrogen in panicle fertiliser and increased density (RPN+ID), and reduced nitrogen in basal fertiliser and increased density (RBN+ID). RPN+ID and RBN+ID, respectively, produced a 3.0% and 5.1% higher yield than CK in both years. The mean grain filling rate (Va) of superior grains in RBN+ID increased by 12.5%, while the mean grain filling rate (Va) of inferior grains in the RPN+ID treatment increased by 4.2% with respect to CK. RPN+ID caused 0.4%, 9.6%, and 13.3% decline in the brown rice rate, chalkiness degree, and chalkiness rate, respectively, while RBN+ID triggered 0.4%, 7.2%, and 11.0% decline in the brown rice rate, chalkiness degree, and chalkiness rate, respectively. RPN+ID stimulated 4.2% and 3.1% increases in flavour and straight-chain amylose values, respectively. Whereas a 20% reduction in basal nitrogen fertiliser and a 32% increase in density improved the yield and appearance quality of rice, a 20% reduction in nitrogen fertiliser at the panicle stage and a 32% increase in density promoted a higher steaming flavour quality. Therefore, an appropriate reduction in nitrogen fertiliser while simultaneously increasing rice density has a significant impact on rice quality, fertiliser pollution reduction, and is a theoretical basis for rice yield and quality improvement in Northeast China. Full article

The way in which alternate wetting and drying irrigation (AWD), as a water-saving practice promoted in rice (Oryza sativa L.) production systems, could enhance the productivity and water use efficiency (WUE) attracts broad attention. This study selected six mid-season indica rice varieties to investigate the impacts of AWD and conventional irrigation (CI) on grain yield, WUE, grain filling, and root traits. A two-year field experiment demonstrated that grain yields and WUE were significantly increased with varietal improvements. With the improvement of varieties, the maximum grain filling rate and mean grain filling rate for both apical superior and basal inferior spikelets were progressively enhanced during the grain filling stage. Compared to CI, AWD significantly enhanced grain yield and WUE. Flag leaf photosynthetic rate and root characteristics, including root weight, root length, root absorbing surface area, root oxidation activity, and zeatin (Z) + zeatin riboside (ZR) contents in panicles, roots, and root bleeding, were superior under AWD across early, mid, and late grain filling stages. Correlation and path analysis showed that improved grain filling in basal inferior spikelets was attributed to delayed root senescence during the grain filling stage under AWD. These results indicated that AWD would be a better irrigation regime to improve yield and WUE by optimizing grain filling and root growth for modern varieties. Full article

Interactions between carbon and nitrogen metabolism are essential for balancing source–sink dynamics in plants. Frequent cold stress disrupts these metabolic processes in rice and reduces grain yield. Two rice cultivars (DN428: cold-tolerant; SJ10: cold-sensitive) were subjected to 19 °C low-temperature stress at full-heading for varying lengths of time to analyze the effects on leaf and grain metabolism. The objective was to track carbon–nitrogen flow and identify factors affecting grain yield. Low-temperature stress significantly reduced the activity of nitrate reductase (NR), glutamine synthetase (GS), glutamate synthase (GOGAT), glutamate dehydrogenase (GDH), glutamic oxaloacetic transaminase (GOT), and glutamic pyruvic transaminase (GPT), in functional leaves compared to the control. This reduction decreased nitrogen accumulation, inhibited chlorophyll synthesis, and slowed photosynthesis. To preserve intracellular osmotic balance and lessen the effects of low temperatures, sucrose, fructose, and total soluble sugar levels, as well as sucrose synthase (SS) and sucrose phosphate synthase (SPS) activities, surged in response to low-temperature stress. However, low-temperature stress significantly reduced the activity of adenosine diphosphate glucose pyrophosphorylase (AGPase), granule-bound starch synthase (GBSS), soluble starch synthase (SSS), and starch branching enzyme (SBE). At the same time, low-temperature stress reduced the area of vascular bundles and phloem, making it difficult to transport carbon and nitrogen metabolites to grains on time. The response of grains to low-temperature stress differs from that of leaves, with prolonged low-temperature exposure causing a gradual decrease in carbon and nitrogen metabolism-related enzyme activities and product accumulation within the grains. The insufficient synthesis of starch precursors and carbon skeletons results in significantly lower thousand-grain weight and seed-setting rates, ultimately contributing to grain yield loss. This decline was more pronounced in inferior grains compared to superior grains. Compared to SJ10, DN428 exhibited higher values across various indicators and smaller declines under low-temperature stress, suggesting enhanced cold-tolerance and a greater capacity to maintain grain yield stability. Full article

To reveal the regulatory effects of nitrogen and phosphorus interactions on grain-filling- and starch-synthesis-related enzymes, and grain weight of superior grains (SGs) and inferior grains (IGs) and taste quality, the japonica rice cultivar Shennong 265 was grown under field conditions with three nitrogen levels (210, 178.5, and 147 kg N ha⁻¹; N3, N2, and N1) and two phosphorus levels (105 and 73.5 kg P ha⁻¹; P2 and P1). At the N3 level, the yield of P1 was significantly lower (by 19.26%) compared to P2; at the N2 and N1 levels, P1 yielded higher than P2, peaking at N2P1. Spikelets per panicle showed P2 exceeding P1 at the same nitrogen level, with the highest for both SGs and IGs observed at N2P2, followed by N2P1. Reductions in nitrogen and phosphorus decreased the grain-filling rate but prolonged the duration for grain-filling. N2P1 maintained grain weight by extending the grain-filling duration across the early, middle, and late stages of IGs, and the middle and late stages of SGs. Increased nitrogen enhanced the activities of soluble starch synthase (SSS) and starch branching enzyme (SBE), whereas increased phosphorus inhibited these activities in SGs but enhanced them in IGs. Reduced nitrogen and phosphorus fertilizer diminished ADP glucose pyrophosphorylase (AGPP) and granule-bound starch synthase (GBSS) activities in SGs and IGs, inhibiting amylose accumulation while enhancing taste value. Compared with N3P2, the taste value of N2P1 increased significantly by 6.93%, attributed to a higher amylopectin/amylose ratio. N2P1 (178.5 kg N ha⁻¹ and 73.5 kg P ha⁻¹) optimized enzyme activity, starch composition, and grain filling, balancing both yield and taste, and thus demonstrated an effective fertilization strategy for stable rice production. Full article

The rapid advancement of deep neural networks has significantly increased demands for computational complexity and data volume. This trend is especially evident with the emergence of large language models, which have rendered traditional architectures such as CPUs and GPGPUs insufficient in meeting performance and energy efficiency requirements. Spatial accelerators present a promising solution by optimizing on-chip compute, storage, and communication resources. In exploring spatial accelerator design spaces, analytical model-based simulators and cycle-accurate simulators are commonly employed, each offering distinct advantages: high computational speed and superior simulation accuracy, respectively. However, the limited accuracy of analytical models and the slow simulation speed of cycle-accurate simulators impede the achievement of globally optimal solutions during design space exploration. Therefore, effectively leveraging the strengths of both simulator types while mitigating their inherent trade-offs is a critical challenge in designing customized spatial accelerators. In this work, we introduce a novel co-exploration methodology that integrates both coarse-grained and fine-grained approaches to navigate design and mapping spaces effectively. We utilize the rapid simulation capabilities of analytical models to perform coarse-grained global exploration, quickly eliminating designs and mapping configurations with inferior performance. Building on the results of this initial exploration, we employ cycle-accurate simulators to conduct fine-grained local exploration within the identified promising regions of the design and mapping spaces. This dual-phase approach aims to identify optimal hardware designs and dataflow mapping strategies that enhance performance and energy efficiency. The experimental results demonstrate that, compared to state-of-the-art methods, our approach reduces the number of exploration points by up to 99%, while achieving a 17.9% reduction in latency, a 2.5% decrease in energy consumption, and a 30.3% improvement in throughput. Full article

It is of great interest to utilize saline fields to promote rice production in China. It has still not been established how salinity stress affects grain-filling characteristics and the relationships with yield formation of rice in a saline field. This experiment was conducted with Ningjing 7 (salinity-tolerant rice variety) and Wuyunjing 30 (salinity-susceptible rice variety) in a non-saline field and a high-saline field in 2021 and 2022. The grain yields of Ningjing 7 and Wuyunjing 30 in a high-saline field were 37.7% and 49.8% lower (p < 0.05) than in a non-saline field across two years. Ningjing 7 exhibited a higher (p < 0.05) grain yield than Wuyunjing 30 in a high-saline field. The reductions in filled-grain percentage and grain weight in inferior grains were greater than in superior grains of Ningjing 7 and Wuyunjing 30. For Ningjing 7 and Wuyunjing 30, the total starch contents in superior and inferior grains at 15, 30, and 45 days after heading were reduced (p < 0.05) in a high-saline field compared to a non-saline field. The ADP–glucose pyrophosphorylase, granule-bound starch synthase, and starch synthase activities after heading in superior and inferior grains in a high-saline field were lower (p < 0.05) than those in a non-saline field, and the reductions were more pronounced for Wuyunjing 30. The maximum grain-filling rate and mean grain-filling rate were decreased, while the time to achieve the maximum grain-filling rate was increased in a high-saline field compared to a non-saline field, especially for Wuyunjing 30. The mean grain-filling rate and grain-filling amount in superior and inferior grains during the early, middle, and late stages were lower in a high-saline field than in a non-saline field. For Ningjing 7 and Wuyunjing 30, the reductions in the grain-filling amount in the inferior grains during the early, middle, and late stages in a high-saline field were greater than those in superior grains. Our results suggest that salinity stress inhibited the grain-filling rate, reduced the total starch content and affected key enzyme activities, which led to the poor sink-filling efficiency and yield performance of rice in a saline field, especially for the salinity-susceptible variety. Full article

Abscisic acid (ABA) plays an important regulatory role in the grain filling process, which in turn will affect the final yield and quality of rice. The ABA biosynthesis genes of OsNCED3 and degradation gene OsABA8ox3 affect the ABA content, and then further regulate the ABA signaling. During the development of rice panicle, compared with primary grains (superior grains) growing on primary branches, secondary grains (inferior grains) growing on secondary branches exhibit characteristics. However, little is reported on the physicochemical characteristics of starch between superior and inferior grains in ABA related transgenic lines. In this study, OsNCED3 and OsABA8ox3 transgenic plants were used as materials. The results showed that compared with the WT, the OsNCED3-RNAi lines on grain weight was consistent with the trend of superior and inferior grains, while the OsABA8ox3-RNAi lines affected superior or inferior grains. The total starch and soluble sugar content of grains decreased in both OsNCED3-RNAi and OsABA8ox3-RNAi lines, and the total starch content of superior and inferior grains in OsABA8ox3-RNAi lines decreased. The starch granule size distribution of all samples showed a bimodal and increased proportion of starch grains with large granule size, in which the influence on inferior grains was greater than that of superior grains, which eventually led to a significant increase in their average granule size. The apparent amylose content of inferior grains increased significantly in most lines. The swelling power of the superior grains decreased significantly, while that of the inferior grains increased significantly. Fourier analysis showed that the order degree of starch granule surface decreased in the superior grains of the RNAi line, while it increased in the inferior grains of the OsABA8ox3-RNAi line but decreased in the OsNCED3-RNAi lines. In the superior grains, the relative crystallinity of starch decreased in the OsNCED3-RNAi lines, but remained unchanged or increased in the OsABA8ox3-RNAi line. In inferior grains, the relative crystallinity of starch decreased in the ABA synthesis RNAi line, but increased in the OsABA8ox3-RNAi line. In summary, the influence of ABA on the physicochemical properties of inferior grains is greater than that of superior grains. Full article

This study investigated the differences in appearance quality among different soft japonica rice varieties based on grain shape, with a particular focus on the broad-ovate soft japonica rice varieties Nanjing 9108 and Nanjing 5718, as well as the slender soft japonica rice varieties Shangshida 19 and Jiahe 218, all sourced from the Yangtze River Delta. The results showed that the slender soft japonica rice varieties exhibited significantly superior appearance quality compared to the broad-ovate varieties. In the case of superior grains, the chalky grain rate of the broad-ovate soft japonica rice was 4307.79 percent higher than that of the slender varieties, and the degree of chalkiness was 8275.00 percent higher. For inferior grains, the chalky grain rate of the broad-ovate soft japonica rice was 238.34 percent higher than the slender varieties, and the degree of chalkiness was 339.96 percent higher. In contrast to the slender soft japonica rice, the broad-ovate varieties had a lower percentage of high-weight grains and a higher percentage of low-weight grains. Compared to the broad-ovate soft japonica rice, the slender varieties exhibited a faster grain-filling rate and shorter effective grain-filling days. Correlation analysis revealed that chalkiness had a significant negative correlation with grain length and aspect ratio. Simultaneously, chalkiness also showed a positive correlation with the number of effective grain-filling days while demonstrating a negative correlation with both the maximum and average grain-filling rates. The slender soft japonica rice exhibited a lower likelihood of developing chalkiness and higher grain-filling efficiency and developed a favorable grain weight distribution. These distinctive attributes significantly contribute to the superior appearance quality of the slender japonica soft rice. Full article

Long-term steel reinforcement corrosion greatly impacts reinforced concrete structures, particularly in marine and coastal settings. Concrete failure leads to human casualties, requiring extensive demolition and maintenance, which represents an inefficient use of energy and resources. This study utilizes microscopic observation, atomic force microscopy (SKPM), electrochemical experiments, and XPS analysis to investigate the corrosion behavior of 500CE and 500E under identical conditions. We compared 500E with 500CE, supplemented with 0.94% Cr, 0.46% Mo, 0.37% Ni, and 0.51% Cu through alloying element regulation to obtain a finer ferrite grain and lower pearlitic content. The results indicate that 500CE maintains a stable potential, whereas 500E exhibits larger grain sizes and significant surface potential fluctuations, which may predispose it to corrosion. In addition, despite its more uniform microstructure and stable electrochemical activity, 500E shows inferior corrosion resistance under prolonged exposure. The electrochemical corrosion rate of 500CE in both the pristine and passivated states and for various passivation durations is slower than that of 500E, indicating superior corrosion performance. Notably, there is a significant increase in the corrosion rate of 500E after 144 h of exposure. This study provides valuable insights into the chloride corrosion phenomena of low-alloy corrosion-resistant steel reinforcement in service, potentially enhancing the longevity of reinforced concrete structures. Full article

Northeast China is an important commercial grain base for China, but also the largest japonica rice production area. However, N, and K fertilizer application and unreasonable application times are prominent contradictions that restrict the development of japonica rice. This study aimed to investigate how to rationally apply N and K fertilizers to affect grain filling and ultimately increase the quality of the rice. In this field study, two N application levels and three K application ratios were set in 2020 and 2021 using Shennong 265 (SN265) and Meifengdao 61 (MF61). We found that the final seed growth and filling rate of SN265 were higher than those of MF61, and their filling characteristics were slow in the early stage and fast in the later stage, with large fluctuations. Appropriate reductions and increases of N and K fertilizer applications, respectively, in the early stage could improve grain filling. Compared with SN265, MF61 had a 3.2% increase in head rice rate, lower amylose and protein content, a decrease of chalkiness degree and chalkiness percentage by 23.96 and 34.00%, respectively, and more reasonable protein components. With the N application increase, the processing quality improved, the amylose and protein content and chalkiness increased, the protein components increased except for the milled rice glutelin, and the rice taste value decreased. At low N levels, increasing the proportion of K application was consistent with the effect of increasing N. The taste value of SN265 decreased linearly with the increase in the ratio of N application to pre-application of K, the highest taste value was obtained when the N fertilizer was applied at a rate of 180 kg ha⁻¹, and the ratio of before and after K fertilizer application was about 1:2. The taste value of MF61 decreased linearly with the N application increase and showed a trend of increasing and then decreasing with the K application increase in the early stage. The taste peak gradually shifted back with the N application increase, and the highest taste value was obtained when N fertilizer was applied at 180 kg ha⁻¹; the ratio of before and after K fertilizer application was about 3:2. By constructing the grain-filling quality evaluation system, the characteristic parameters of superior and inferior grains at the early and late stages of grain filling, respectively, greatly affected the rice taste value. Additionally, the percentage of the rice grain weight at the maximum grain-filling rate to the final rice grain weight (I) of superior grains, glutelin content, and value of the RVA profile characteristics were all critical reference indicators for rice taste quality. Full article

Polyamines and ethylene are key regulators of the growth and development, quality formation, and stress response of cereal crops such as rice. However, it remains unclear whether the application of these regulators could improve the nutritional quality via increasing amino acids in rice grains. This study examined the role of exogenous polyamines and ethylene in regulating amino acid levels in the milled rice of earlier-flowered superior grain (SG) and later-flowered inferior grain (IG). Two rice varieties were field grown, and either 1 mmol L⁻¹ spermidine (Spd) or 50 μmol L⁻¹ amino-ethoxyvinylglycine (AVG) was applied to panicles at the early grain-filling stage. The control check (CK) was applied with deionized water. The results showed that the Spd or AVG applications significantly increased polyamine (spermine (Spm) and Spd) contents and decreased ethylene levels in both SG and IG and significantly increased amino acid levels in the milled rice of SG and IG relative to the CK. Collectively, the application of Spd or AVG can increase amino acid-based nutritional quality and grain yield via increasing polyamine (Spm and Spd) contents and reducing ethylene levels in both SG and IG of rice. Full article

This work investigated the effect of Fe/Mn ratio on the microstructure and mechanical properties of non-equimolar Fe_80−xMn_xCo₁₀Cr₁₀ (x = 30% and 50%) high-entropy alloys (HEAs) fabricated by laser powder bed fusion (LPBF) additive manufacturing. Process optimization was conducted to achieve fully dense Fe₃₀Mn₅₀Co₁₀Cr₁₀ and Fe₅₀Mn₃₀Co₁₀Cr₁₀ HEAs using a volumetric energy density of 105.82 J·mm⁻³. The LPBF-printed Fe₃₀Mn₅₀Co₁₀Cr₁₀ HEA exhibited a single face-centered cubic (FCC) phase, while the Fe₅₀Mn₃₀Co₁₀Cr₁₀ HEA featured a hexagonal close-packed (HCP) phase within the FCC matrix. Notably, the fraction of HCP phase in the Fe₅₀Mn₃₀Co₁₀Cr₁₀ HEAs increased from 0.94 to 28.10%, with the deformation strain ranging from 0 to 20%. The single-phase Fe₃₀Mn₅₀Co₁₀Cr₁₀ HEA demonstrated a remarkable combination of high yield strength (580.65 MPa) and elongation (32.5%), which surpassed those achieved in the FeMnCoCr HEA system. Comparatively, the dual-phase Fe₅₀Mn₃₀Co₁₀Cr₁₀ HEA exhibited inferior yield strength (487.60 MPa) and elongation (22.3%). However, it displayed superior ultimate tensile strength (744.90 MPa) compared to that in the Fe₃₀Mn₅₀Co₁₀Cr₁₀ HEA (687.70 MPa). The presence of FCC/HCP interfaces obtained in the Fe₅₀Mn₃₀Co₁₀Cr₁₀ HEA resulted in stress concentration and crack expansion, thereby leading to reduced ductility but enhanced resistance against grain slip deformation. Consequently, these interfaces facilitated an earlier attainment of yield limit point and contributed to increased ultimate tensile strength in the Fe₅₀Mn₃₀Co₁₀Cr₁₀ HEA. These findings provide valuable insights into the microstructure evolution and mechanical behavior of LPBF-printed metastable FeMnCoCr HEAs. Full article

Increasing grain weight under dense planting conditions can further improve maize yield. 6-BA is known to be involved in regulating grain development and influencing grain weight. Maize grain development is closely linked to starch accumulation and hormone levels. In this work, the effects of applying 6-BA at the flowering stage under high density on the grain filling characteristics, starch content, starch synthesis critical enzyme activity, and endogenous hormones levels of maize grains (including inferior grains (IGs) and superior grains (SGs)) of two high-yielding summer maize varieties widely cultivated in China were investigated. The findings indicated that applying 6-BA significantly improved maize yield compared to the control, mainly as a result of increased grain weight due to a faster grain filling rate. Additionally, the activities of enzymes associated with starch synthesis, including sucrose synthase (SuSy), ADP-glucose pyrophosphorylase (AGPase), granule-bound starch synthase (GBSS), soluble starch synthase (SSS), and starch branching enzyme (SBE), were all increased following 6-BA application, thus facilitating starch accumulation in the grains. Applying 6-BA also increased the zeatin riboside (ZR), indole-3-acetic acid (IAA), and abscisic acid (ABA) levels, and reduced the gibberellin (GA₃) level in the grains, which further improved grain filling. It is worth noting that IG had a poorer filling process than SG, possibly due to the low activities of critical enzymes for starch synthesis and imbalanced endogenous hormones levels. However, IG responded more strongly to exogenous 6-BA than SG. It appears that applying 6-BA is beneficial in improving filling characteristics, promoting starch accumulation by enhancing the activities of critical enzymes for starch synthesis, and altering endogenous hormones levels in the grains, thus improving grain filling and increasing the final grain weight and yield of maize grown under crowded conditions. These results provide theoretical and technical support for the further utilization of exogenous hormones in high-density maize production. Full article

The prediction of crop yield plays a crucial role in national economic development, encompassing grain storage, processing, and grain price trends. Employing multiple sensors to acquire remote sensing data and utilizing machine learning algorithms can enable accurate, fast, and nondestructive yield prediction for maize crops. However, current research heavily relies on single-type remote sensing data and traditional machine learning methods, resulting in the limited robustness of yield prediction models. To address these limitations, this study introduces a field-scale maize yield prediction model named the convolutional neural network–attention–long short-term memory network (CNN-attention-LSTM) model, which utilizes multimodal remote sensing data collected by multispectral and light detection and ranging (LIDAR) sensors mounted on unmanned aerial vehicles (UAVs). The model incorporates meteorological data throughout the crop reproductive stages and employs the normalized difference vegetation index (NDVI), normalized difference red edge (NDRE), soil-adjusted vegetation index (SAVI), and enhanced vegetation index (EVI) for the initial part of the vegetative stage (initial part of the V period), the later part of the vegetative stage (later part of the V period), the reproductive stage (R period), and the maturity stage (M period), along with LIDAR data for Point_75–100 in the later part of the V period, Point_80–100 in the R period, and Point_50–100 in the M period, complemented by corresponding meteorological data as inputs. The resulting yield estimation demonstrates exceptional performance, with an R² value of 0.78 and an rRMSE of 8.27%. These results surpass previous research and validate the effectiveness of multimodal data in enhancing yield prediction models. Furthermore, to assess the superiority of the proposed model, four machine learning algorithms—multiple linear regression (MLR), random forest regression (RF), support vector machine (SVM), and backpropagation (BP)—are compared to the CNN-attention-LSTM model through experimental analysis. The outcomes indicate that all alternative models exhibit inferior prediction accuracy compared to the CNN-attention-LSTM model. Across the test dataset within the study area, the R² values for various nitrogen fertilizer levels consistently exceed 0.75, illustrating the robustness of the proposed model. This study introduces a novel approach for assessing maize crop yield and provides valuable insights for estimating the yield of other crops. Full article