Stomatal, mesophyll and biochemical limitations to photosynthesis and their relationship with leaf structure over an elevation gradient in two conifers.

Photosynthetic responses across complex elevational gradients provides insight into fundamental processes driving responses of plant growth and net primary production to environmental change. Gas exchange of needles and twig water potential were measured in two widespread coniferous tree species, Pinus contorta and Picea engelmannii, over an 800-m elevation gradient in southeastern Wyoming, USA. We hypothesized that limitations to photosynthesis imposed by mesophyll conductance (gm) would be greatest at the highest elevation sites due to higher leaf mass per area (LMA) and that estimations of maximum rate of carboxylation (Vcmax) without including gm would obscure elevational patterns of photosynthetic capacity. We found that gm decreased with elevation for P. contorta and remained constant for P. engelmannii, but in general, limitation to photosynthesis by gm was small. Indeed, estimations of Vcmax when including gm were equivalent to those estimated without including gm and no correlation was found between gm and LMA nor between gm and leaf N. Stomatal conductance (gs) and biochemical demand for CO2 were by far the most limiting processes to photosynthesis at all sites along the elevation gradient. Photosynthetic capacity (A) and gs were influenced strongly by differences in soil water availability across the elevation transect, while gm was less responsive to water availability. Based on our analysis, variation in gm plays only a minor role in driving patterns of photosynthesis in P. contorta and P. engelmannii across complex elevational gradients in dry, continental environments of the Rocky Mountains and accurate modeling of photosynthesis, growth and net primary production in these forests may not require detailed estimation of this trait value.

Physio-biochemical, agronomical, and gene expression analysis reveals different responsive approach to low nitrogen in contrasting rice cultivars for nitrogen use efficiency.

BACKGROUND: Nitrogen (N) is an essential macronutrient for plant growth and development as it is an essential constituent of biomolecules. Its availability directly impacts crop yield. Increased N application in crop fields has caused environmental and health problems, and decreasing nitrogen inputs are in demand to maintain crop production sustainability. Understanding the molecular mechanism of N utilization could play a crucial role in improving the nitrogen use efficiency (NUE) of crop plants. METHODS AND RESULTS: In the present study, the effect of low N supply on plant growth, physio-biochemical, chlorophyll fluorescence attributes, yield components, and gene expression analysis were measured at six developmental stages in rice cultivars. Two rice cultivars were grown with a supply of optimium (120 kg ha-1) and low N (60 kg ha-1). Cultivar Vikramarya excelled Aditya at low N supply, and exhibits enhanced plant growth, physiological efficiency, agronomic efficiency, and improved NUE due to higher N uptake and utilization at low N treatment. Moreover, plant biomass, leaf area, and photosynthetic rate were significantly higher in cv. Vikramarya than cv. Aditya at different growth stages, under low N treatment. In addition, enzymatic activities in cultivar Vikramarya were higher than cultivar Aditya under low nitrogen, indicating its greater potential for N metabolism. Gene expression analysis was carried out for the most important nitrogen assimilatory enzymes, such as nitrate reductase (NR), nitrite reductase (NiR), glutamine synthetase (GS), and glutamate synthase (GOGAT). Expression levels of these genes at different growth stages were significantly higher in cv. Vikramarya compared to cv. Aditya at low N supply. Our findings suggest that improving NUE needs specific revision in N metabolism and physiological assimilation. CONCLUSION: Overall differences in plant growth, physiological efficiency, biochemical activities, and expression levels of N metabolism genes in N-efficient and N-inefficient rice cultivars need a specific adaptation to N metabolism. Regulatory genes may separately or in conjunction, enhance the NUE. These results provide a platform for selecting crop cultivars for nitrogen utilization efficiency at low N treatment.

Can evolutionary history predict plant plastic responses to climate change?

Plant plastic responses are critical to the adaptation and survival of species under climate change, but whether they are constrained by evolutionary history (phylogeny) is largely unclear. Plant leaf traits are key in determining plants' performance in different environments, and if these traits and their variation are phylogenetically dependent, predictions could be made to identify species vulnerable to climate change. We compiled data on three leaf traits (photosynthetic rate, specific leaf area, and leaf nitrogen content) and their variation under four environmental change scenarios (warming, drought, elevated CO2 , or nitrogen addition) for 434 species, from 210 manipulation experiments. We found phylogenetic signal in the three traits but not in their variation under the four scenarios. This indicates that closely related species show similar traits but that their plastic responses could not be predicted from species relatedness under environmental change. Meanwhile, phylogeny weakened the slopes but did not change the directions of conventional pairwise trait relationships, suggesting that co-evolved leaf trait pairs have consistent responses under contrasting environmental conditions. Phylogeny can identify lineages rich in species showing similar traits and predict their relationships under climate change, but the degree of plant phenotypic variation does not vary consistently across evolutionary clades.

Predicting biochemical acclimation of leaf photosynthesis in soybean under in-field canopy warming using hyperspectral reflectance.

Traditional gas exchange measurements are cumbersome, which makes it difficult to capture variation in biochemical parameters, namely the maximum rate of carboxylation measured at a reference temperature (Vcmax25 ) and the maximum electron transport at a reference temperature (Jmax25 ), in response to growth temperature over time from days to weeks. Hyperspectral reflectance provides reliable measures of Vcmax25 and Jmax25 ; however, the capability of this method to capture biochemical acclimations of the two parameters to high growth temperature over time has not been demonstrated. In this study, Vcmax25 and Jmax25 were measured over multiple growth stages during two growing seasons for field-grown soybeans using both gas exchange techniques and leaf spectral reflectance under ambient and four elevated canopy temperature treatments (ambient+1.5, +3, +4.5, and +6°C). Spectral vegetation indices and machine learning methods were used to build predictive models for Vcmax25 and Jmax25 , based on the leaf reflectance. Results showed that these models yielded an R2 of 0.57-0.65 and 0.48-0.58 for Vcmax25 and Jmax25 , respectively. Hyperspectral reflectance captured biochemical acclimation of leaf photosynthesis to high temperature in the field, improving spatial and temporal resolution in the ability to assess the impact of future warming on crop productivity.

Estimation of Leaf Nitrogen Content in Wheat Based on Fusion of Spectral Features and Deep Features from Near Infrared Hyperspectral Imagery.

Nitrogen is an important indicator for monitoring wheat growth. The rapid development and wide application of non-destructive detection provide many approaches for estimating leaf nitrogen content (LNC) in wheat. Previous studies have shown that better results have been obtained in the estimation of LNC in wheat based on spectral features. However, the lack of automatically extracted features leads to poor universality of the estimation model. Therefore, a feature fusion method for estimating LNC in wheat by combining spectral features with deep features (spatial features) was proposed. The deep features were automatically obtained with a convolutional neural network model based on the PyTorch framework. The spectral features were obtained using spectral information including position features (PFs) and vegetation indices (VIs). Different models based on feature combination for evaluating LNC in wheat were constructed: partial least squares regression (PLS), gradient boosting decision tree (GBDT), and support vector regression (SVR). The results indicate that the model based on the fusion feature from near-ground hyperspectral imagery has good estimation effect. In particular, the estimation accuracy of the GBDT model is the best (R2 = 0.975 for calibration set, R2 = 0.861 for validation set). These findings demonstrate that the approach proposed in this study improved the estimation performance of LNC in wheat, which could provide technical support in wheat growth monitoring.

The acclimation of leaf photosynthesis of wheat and rice to seasonal temperature changes in T-FACE environments.

Crops show considerable capacity to adjust their photosynthetic characteristics to seasonal changes in temperature. However, how photosynthesis acclimates to changes in seasonal temperature under future climate conditions has not been revealed. We measured leaf photosynthesis (An ) of wheat (Triticum aestivum L.) and rice (Oryza sativa L.) grown under four combinations of two levels of CO2 (ambient and enriched up to 500 µmol/mol) and two levels of canopy temperature (ambient and increased by 1.5-2.0°C) in temperature by free-air CO2 enrichment (T-FACE) systems. Parameters of a biochemical C3 -photosynthesis model and of a stomatal conductance (gs ) model were estimated for the four conditions and for several crop stages. Some biochemical parameters related to electron transport and most gs parameters showed acclimation to seasonal growth temperature in both crops. The acclimation response did not differ much between wheat and rice, nor among the four treatments of the T-FACE systems, when the difference in the seasonal growth temperature was accounted for. The relationships between biochemical parameters and leaf nitrogen content were consistent across leaf ranks, developmental stages, and treatment conditions. The acclimation had a strong impact on gs model parameters: when parameter values of a particular stage were used, the model failed to correctly estimate gs values of other stages. Further analysis using the coupled gs -biochemical photosynthesis model showed that ignoring the acclimation effect did not result in critical errors in estimating leaf photosynthesis under future climate, as long as parameter values were measured or derived from data obtained before flowering.

Short-term influence of biochar on soil temperature, liquid moisture content and soybean growth in a seasonal frozen soil area.

In the past decade, the application of biochar in agricultural soils has attracted wide attention. However, few studies have carefully explored the effects of biochar modification on plant leaf nutrients and the physiological process of plant leaves. To provide a better growing environment for crops and explore the best regulation mode of biochar in the farmland soil environment in the typical black loam area of Heilongjiang Province, through field experiments, we selected soybeans as the test crop and applied biochar in the soil. The agronomic characteristics and soil conditions of soybean plants were monitored by stage. The effects of different application methods and biochar gradients on the water and heat changes in soil tillage layers during different growth stages of crops were discussed, and the subtle differences of agronomic characteristics in different growth stages of crops were compared. The results showed that all kinds of biochar application modes could not change the general trend of water and heat change in soil tillage layer affected by environmental factors, and the effect of biochar application on soil liquid moisture content at 20 cm soil layer was not obvious. Biochar application can increase plant height and reduce stem diameter, but the effect is non-linear. The leaf nitrogen content (Leaf N-content) and leaf chlorophyll relative content (SPAD) were vertically distributed in the canopy, but they did not change significantly with the change of biochar application rate and mode. The application of biochar in autumn may bring crops into maturity earlier. Under the biochar application rate of 9 kg m-2, the mixed application in spring and autumn can bring the best biochar application effect.

Qualitative and quantitative diagnosis of nitrogen nutrition of tea plants under field condition using hyperspectral imaging coupled with chemometrics.

BACKGROUND: Rapid and accurate diagnosis of nitrogen (N) status in field crops is of great significance for site-specific N fertilizer management. This study aimed to evaluate the potential of hyperspectral imaging coupled with chemometrics for the qualitative and quantitative diagnosis of N status in tea plants under field conditions. RESULTS: Hyperspectral data from mature leaves of tea plants with different N application rates were preprocessed by standard normal variate (SNV). Partial least squares discriminative analysis (PLS-DA) and least squares-support vector machines (LS-SVM) were used for the classification of different N status. Furthermore, partial least squares regression (PLSR) was used for the prediction of N content. The results showed that the LS-SVM model yielded better performance with correct classification rates of 82% and 92% in prediction sets for the diagnosis of different N application rates and N status, respectively. The PLSR model for leaf N content (LNC) showed excellent performance, with correlation coefficients of 0.924, root mean square error of 0.209, and residual predictive deviation of 2.686 in the prediction set. In addition, the important wavebands of the PLSR model were interpreted based on regression coefficients. CONCLUSION: Overall, our results suggest that the hyperspectral imaging technique can be an effective and accurate tool for qualitative and quantitative diagnosis of N status in tea plants. © 2019 Society of Chemical Industry.

Symbiotic dinitrogen fixation is seasonal and strongly regulated in water-limited environments.

Plants, especially perennials, growing in drylands and seasonally dry ecosystems are uniquely adapted to dry conditions. Legume shrubs and trees, capable of symbiotic dinitrogen (N2 ) fixation, often dominate in drylands. However, the strategies that allow symbiotic fixation in these ecosystems, and their influence on the nitrogen cycle, are largely unresolved. We evaluated the climatic, biogeochemical and ontogenetic factors influencing nitrogen fixation in an abundant Mediterranean legume shrub, Calicotome villosa. We measured nodulation, fixation rate, nitrogen allocation and soil biogeochemistry in three field sites over a full year. A controlled experiment evaluated differences in plant regulation of fixation as a function of soil nutrient availability and seedling and adult developmental stages. We found a strong seasonal pattern, shifting between high fixation rates during the rainy season at flowering and seed-set times to almost none in the rainless season. Under controlled conditions, plants downregulated fixation in response to soil nitrogen availability, but this response was stronger in seedlings than in adult shrubs. Finally, we did not find elevated soil nitrogen under N2 -fixing shrubs. We conclude that seasonal nitrogen fixation, regulation of fixation, and nitrogen conservation are key adaptations influencing the dominance of dryland legumes in the community, with broader consequences on the ecosystem nitrogen cycle.

Hyperspectral-based Estimation of Leaf Nitrogen Content in Corn Using Optimal Selection of Multiple Spectral Variables.

Accurate and dynamic monitoring of crop nitrogen status is the basis of scientific decisions regarding fertilization. In this study, we compared and analyzed three types of spectral variables: Sensitive spectral bands, the position of spectral features, and typical hyperspectral vegetation indices. First, the Savitzky-Golay technique was used to smooth the original spectrum, following which three types of spectral parameters describing crop spectral characteristics were extracted. Next, the successive projections algorithm (SPA) was adopted to screen out the sensitive variable set from each type of parameters. Finally, partial least squares (PLS) regression and random forest (RF) algorithms were used to comprehensively compare and analyze the performance of different types of spectral variables for estimating corn leaf nitrogen content (LNC). The results show that the integrated variable set composed of the optimal ones screened by SPA from three types of variables had the best performance for LNC estimation by the validation data set, with the values of R2, root means square error (RMSE), and normalized root mean square error (NRMSE) of 0.77, 0.31, and 17.1%, and 0.55, 0.43, and 23.9% from PLS and RF, respectively. It indicates that the PLS model with optimally multitype spectral variables can provide better fits and be a more effective tool for evaluating corn LNC.

Within-species patterns challenge our understanding of the leaf economics spectrum.

The utility of plant functional traits for predictive ecology relies on our ability to interpret trait variation across multiple taxonomic and ecological scales. Using extensive data sets of trait variation within species, across species and across communities, we analysed whether and at what scales leaf economics spectrum (LES) traits show predicted trait-trait covariation. We found that most variation in LES traits is often, but not universally, at high taxonomic levels (between families or genera in a family). However, we found that trait covariation shows distinct taxonomic scale dependence, with some trait correlations showing opposite signs within vs. across species. LES traits responded independently to environmental gradients within species, with few shared environmental responses across traits or across scales. We conclude that, at small taxonomic scales, plasticity may obscure or reverse the broad evolutionary linkages between leaf traits, meaning that variation in LES traits cannot always be interpreted as differences in resource use strategy.

Do all leaf photosynthesis parameters of rice acclimate to elevated CO2 , elevated temperature, and their combination, in FACE environments?

Leaf photosynthesis of crops acclimates to elevated CO2 and temperature, but studies quantifying responses of leaf photosynthetic parameters to combined CO2 and temperature increases under field conditions are scarce. We measured leaf photosynthesis of rice cultivars Changyou 5 and Nanjing 9108 grown in two free-air CO2 enrichment (FACE) systems, respectively, installed in paddy fields. Each FACE system had four combinations of two levels of CO2 (ambient and enriched) and two levels of canopy temperature (no warming and warmed by 1.0-2.0°C). Parameters of the C3 photosynthesis model of Farquhar, von Caemmerer and Berry (the FvCB model), and of a stomatal conductance (gs ) model were estimated for the four conditions. Most photosynthetic parameters acclimated to elevated CO2 , elevated temperature, and their combination. The combination of elevated CO2 and temperature changed the functional relationships between biochemical parameters and leaf nitrogen content for Changyou 5. The gs model significantly underestimated gs under the combination of elevated CO2 and temperature by 19% for Changyou 5 and by 10% for Nanjing 9108 if no acclimation was assumed. However, our further analysis applying the coupled gs -FvCB model to an independent, previously published FACE experiment showed that including such an acclimation response of gs hardly improved prediction of leaf photosynthesis under the four combinations of CO2 and temperature. Therefore, the typical procedure that crop models using the FvCB and gs models are parameterized from plants grown under current ambient conditions may not result in critical errors in projecting productivity of paddy rice under future global change.

Invasive species' leaf traits and dissimilarity from natives shape their impact on nitrogen cycling: a meta-analysis.

Many exotic species have little apparent impact on ecosystem processes, whereas others have dramatic consequences for human and ecosystem health. There is growing evidence that invasions foster eutrophication. We need to identify species that are harmful and systems that are vulnerable to anticipate these consequences. Species' traits may provide the necessary insights. We conducted a global meta-analysis to determine whether plant leaf and litter functional traits, and particularly leaf and litter nitrogen (N) content and carbon: nitrogen (C : N) ratio, explain variation in invasive species' impacts on soil N cycling. Dissimilarity in leaf and litter traits among invaded and noninvaded plant communities control the magnitude and direction of invasion impacts on N cycling. Invasions that caused the greatest increases in soil inorganic N and mineralization rates had a much greater litter N content and lower litter C : N in the invaded than the reference community. Trait dissimilarities were better predictors than the trait values of invasive species alone. Quantifying baseline community tissue traits, in addition to those of the invasive species, is critical to understanding the impacts of invasion on soil N cycling.

Leaf photosynthetic rate and mesophyll cell anatomy changes during ontogenesis in backcrossed indica × japonica rice inbred lines.

The high-yielding indica rice variety, 'Takanari', has the high rate of leaf photosynthesis compared with the commercial japonica varieties. Among backcrossed inbred lines from a cross between 'Takanari' and a japonica variety, 'Koshihikari', two lines, BTK-a and BTK-b, showed approximately 20% higher photosynthetic rate than that of 'Takanari' for a flag leaf at full heading. This is a highest recorded rate of rice leaf photosynthesis. Here, the timing and cause of the increased leaf photosynthesis in the BTK lines were investigated by examining the photosynthesis and related parameters, as well as mesophyll cell anatomy during ontogenesis. Their photosynthetic rate was greater than that of 'Takanari' in the 13th leaf, as well as the flag leaf, but there were no differences in the 7th and 10th leaves. There were no consistent differences in the stomatal conductance, or the leaf nitrogen and Rubisco contents in the 13th and flag leaves. The total surface area of mesophyll cells per leaf area (TAmes) in the 13th and flag leaves increased significantly in the BTK lines due to the increased number and developed lobes of mesophyll cells compared with in 'Takanari'. The mesophyll conductance (g m) became greater in the BTK lines compared with 'Takanari' in the flag leaves but not in the 10th leaves. A close correlation was observed between TAmes and g m. We concluded that the increased mesophyll conductance through the development of mesophyll cells during the reproductive period is a probable cause of the greater photosynthetic rate in the BTK lines.

Inter-genotypic differences in drought tolerance of maritime pine are modified by elevated [CO2].

Background and Aims: Despite the importance of growth [CO 2 ] and water availability for tree growth and survival, little information is available on how the interplay of these two factors can shape intraspecific patterns of functional variation in tree species, particularly for conifers. The main objective of the study was to test whether the range of realized drought tolerance within the species can be affected by elevated [CO 2 ]. Methods: Intraspecific variability in leaf gas exchange, growth rate and other leaf functional traits were studied in clones of maritime pine. A factorial experiment including water availability, growth [CO 2 ] and four different genotypes was conducted in growth rooms. A 'water deficit' treatment was imposed by applying a cycle of progressive soil water depletion and recovery at two levels of growth [CO 2 ]: 'ambient [CO 2 ]' (aCO 2 400 µmol mol -1 ) and 'elevated [CO 2 ]' (eCO 2 800 µmol mol -1 ). Key Results: eCO2 had a neutral effect on the impact of drought on growth and leaf gas exchange of the most drought-sensitive genotypes while it aggravated the impact of drought on the most drought-tolerant genotypes at aCO2. Thus, eCO2 attenuated genotypic differences in drought tolerance as compared with those observed at aCO2. Genotypic variation at both levels of growth [CO2] was found in specific leaf area and leaf nitrogen content but not in other physiological leaf traits such as intrinsic water use efficiency and leaf osmotic potential. eCO2 increased Δ 13 C but had no significant effect on δ 18 O. This effect did not interact with the impact of drought, which increased δ 18 O and decreased Δ 13 C. Nevertheless, correlations between Δ 13 C and δ 18 O indicated the non-stomatal component of water use efficiency in this species can be particularly sensitive to drought. Conclusions: Evidence from this study suggests elevated [CO 2 ] can modify current ranges of drought tolerance within tree species.

Relationships between the timing of budburst, plant traits, and distribution of 24 coexisting woody species in a warm-temperate forest in Japan.

PREMISE OF THE STUDY: Timing of budburst (DBB) may be related to the functional traits and distributions of woody species in temperate regions. Although many previous studies have investigated DBB in a number of temperate species, it has seldom been linked to multiple plant trait relationships. METHODS: DBB and plant traits were investigated for 24 woody species for 2 years in a warm-temperate secondary forest in Japan. Particular attention was paid to differences in trait relationships between coexisting deciduous and evergreen broad-leaved species. KEY RESULTS: DBB was correlated with plant traits in deciduous but not evergreen broad-leaved species; DBB was later for deciduous species with greater leaf mass, leaf area, vessel diameter, and leaf nitrogen content per unit mass. In addition, DBB was later for species with more northern distributions in deciduous and evergreen species. CONCLUSIONS: Clear differences in the trait relationships between deciduous and evergreen broad-leaved species might be caused by different selection pressures on DBB; selection is expected to be more severe in deciduous species. Overall, the continuous variable of vessel diameter might be used as a simple and effective trait to predict DBB of deciduous species regardless of wood anatomy; however, no such traits were detected as effective predictors of DBB in evergreen species at this study site. In addition, DBB was earlier for the species of more southern distributions, suggesting that such species benefit more from warming.

Evolutionary history underlies plant physiological responses to global change since the last glacial maximum.

Assessing family- and species-level variation in physiological responses to global change across geologic time is critical for understanding factors that underlie changes in species distributions and community composition. Here, we used stable carbon isotopes, leaf nitrogen content and stomatal measurements to assess changes in leaf-level physiology in a mixed conifer community that underwent significant changes in composition since the last glacial maximum (LGM) (21 kyr BP). Our results indicate that most plant taxa decreased stomatal conductance and/or maximum photosynthetic capacity in response to changing conditions since the LGM. However, plant families and species differed in the timing and magnitude of these physiological responses, and responses were more similar within families than within co-occurring species assemblages. This suggests that adaptation at the level of leaf physiology may not be the main determinant of shifts in community composition, and that plant evolutionary history may drive physiological adaptation to global change over recent geologic time.

Introgression of two chromosome regions for leaf photosynthesis from an indica rice into the genetic background of a japonica rice.

Increases in rates of individual leaf photosynthesis (P n) are critical for future increases of rice yields. A previous study, using introgression lines derived from a cross between indica cultivar Habataki, with one of the highest recorded values of P n, and the Japanese elite cultivar Koshihikari, identified four QTLs (qCAR4, qCAR5, qCAR8, and qCAR11) that affect P n. The present study examined the combined effect of qCAR4 and qCAR8 on P n in the genetic background of Koshihikari. The pyramided near-isogenic line NIL(qCAR4+qCAR8) showed higher P n than both NIL(qCAR4) and NIL(qCAR8), equivalent to that of Habataki despite being due to only two out of the four QTLs. The high P n of NIL(qCAR4+qCAR8) may be attributable to the high leaf nitrogen content, which may have been inherited from NIL(qCAR4), to the large hydraulic conductance due to the large root surface area from NIL(qCAR4), and to the high hydraulic conductivity from NIL(qCAR8). It might be also attributable to high mesophyll conductance, which may have been inherited from NIL(qCAR4). The induction of mesophyll conductance and the high leaf nitrogen content and high hydraulic conductivity could not be explained in isolation from the Koshihikari background. These results suggest that QTL pyramiding is a useful approach in rice breeding aimed at increasing P n.

Monitoring of Nitrogen Concentration in Soybean Leaves at Multiple Spatial Vertical Scales Based on Spectral Parameters.

Nitrogen is a fundamental component for building amino acids and proteins, playing a crucial role in the growth and development of plants. Leaf nitrogen concentration (LNC) serves as a key indicator for assessing plant growth and development. Monitoring LNC provides insights into the absorption and utilization of nitrogen from the soil, offering valuable information for rational nutrient management. This, in turn, contributes to optimizing nutrient supply, enhancing crop yields, and minimizing adverse environmental impacts. Efficient and non-destructive estimation of crop LNC is of paramount importance for on-field crop management. Spectral technology, with its advantages of repeatability and high-throughput observations, provides a feasible method for obtaining LNC data. This study explores the responsiveness of spectral parameters to soybean LNC at different vertical scales, aiming to refine nitrogen management in soybeans. This research collected hyperspectral reflectance data and LNC data from different leaf layers of soybeans. Three types of spectral parameters, nitrogen-sensitive empirical spectral indices, randomly combined dual-band spectral indices, and "three-edge" parameters, were calculated. Four optimal spectral index selection strategies were constructed based on the correlation coefficients between the spectral parameters and LNC for each leaf layer. These strategies included empirical spectral index combinations (Combination 1), randomly combined dual-band spectral index combinations (Combination 2), "three-edge" parameter combinations (Combination 3), and a mixed combination (Combination 4). Subsequently, these four combinations were used as input variables to build LNC estimation models for soybeans at different vertical scales using partial least squares regression (PLSR), random forest (RF), and a backpropagation neural network (BPNN). The results demonstrated that the correlation coefficients between the LNC and spectral parameters reached the highest values in the upper soybean leaves, with most parameters showing significant correlations with the LNC (p < 0.05). Notably, the reciprocal difference index (VI6) exhibited the highest correlation with the upper-layer LNC at 0.732, with a wavelength combination of 841 nm and 842 nm. In constructing the LNC estimation models for soybeans at different leaf layers, the accuracy of the models gradually improved with the increasing height of the soybean plants. The upper layer exhibited the best estimation performance, with a validation set coefficient of determination (R2) that was higher by 9.9% to 16.0% compared to other layers. RF demonstrated the highest accuracy in estimating the upper-layer LNC, with a validation set R2 higher by 6.2% to 8.8% compared to other models. The RMSE was lower by 2.1% to 7.0%, and the MRE was lower by 4.7% to 5.6% compared to other models. Among different input combinations, Combination 4 achieved the highest accuracy, with a validation set R2 higher by 2.3% to 13.7%. In conclusion, by employing Combination 4 as the input, the RF model achieved the optimal estimation results for the upper-layer LNC, with a validation set R2 of 0.856, RMSE of 0.551, and MRE of 10.405%. The findings of this study provide technical support for remote sensing monitoring of soybean LNCs at different spatial scales.

Using Machine Learning Methods Combined with Vegetation Indices and Growth Indicators to Predict Seed Yield of Bromus inermis.

Smooth bromegrass (Bromus inermis) is a perennial, high-quality forage grass. However, its seed yield is influenced by agronomic practices, climatic conditions, and the growing year. The rapid and effective prediction of seed yield can assist growers in making informed production decisions and reducing agricultural risks. Our field trial design followed a completely randomized block design with four blocks and three nitrogen levels (0, 100, and 200 kg·N·ha-1) during 2022 and 2023. Data on the remote vegetation index (RVI), the normalized difference vegetation index (NDVI), the leaf nitrogen content (LNC), and the leaf area index (LAI) were collected at heading, anthesis, and milk stages. Multiple linear regression (MLR), support vector machine (SVM), and random forest (RF) regression models were utilized to predict seed yield. In 2022, the results indicated that nitrogen application provided a sufficiently large range of variation of seed yield (ranging from 45.79 to 379.45 kg ha⁻¹). Correlation analysis showed that the indices of the RVI, the NDVI, the LNC, and the LAI in 2022 presented significant positive correlation with seed yield, and the highest correlation coefficient was observed at the heading stage. The data from 2022 were utilized to formulate a predictive model for seed yield. The results suggested that utilizing data from the heading stage produced the best prediction performance. SVM and RF outperformed MLR in prediction, with RF demonstrating the highest performance (R2 = 0.75, RMSE = 51.93 kg ha-1, MAE = 29.43 kg ha-1, and MAPE = 0.17). Notably, the accuracy of predicting seed yield for the year 2023 using this model had decreased. Feature importance analysis of the RF model revealed that LNC was a crucial indicator for predicting smooth bromegrass seed yield. Further studies with an expanded dataset and integration of weather data are needed to improve the accuracy and generalizability of the model and adaptability for the growing year.