Adaboost-Based Machine Learning Improved the Modeling Robust and Estimation Accuracy of Pear Leaf Nitrogen Concentration by In-Field VIS-NIR Spectroscopy.
Ontology highlight
ABSTRACT:
SUBMITTER: Wang J
PROVIDER: S-EPMC8473462 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Soil organic carbon (SOC) is an important soil property that has profound impact on soil quality and plant growth. With 140 soil samples collected from Ebinur Lake Wetland National Nature Reserve, Xinjiang Uyghur Autonomous Region of China, this research evaluated the feasibility of visible/near infrared (VIS/NIR) spectroscopy data (350-2,500 nm) and simulated EO-1 Hyperion data to estimate SOC in arid wetland regions. Three machine learning algorithms including Ant Colony Optimization-interval Partial Least Squares (ACO-iPLS), Recursive Feature Elimination-Support Vector Machine (RF-SVM), and Random Forest (RF) were employed to select spectral features and further estimate SOC. Results indicated that the feature wavelengths pertaining to SOC were mainly within the ranges of 745-910 nm and 1,911-2,254 nm. The combination of RF-SVM and first derivative pre-processing produced the highest estimation accuracy with the optimal values of Rt (correlation coefficient of testing set), RMSE t and RPD of 0.91, 0.27% and 2.41, respectively. The simulated EO-1 Hyperion data combined with Support Vector Machine (SVM) based recursive feature elimination algorithm produced the most accurate estimate of SOC content. For the testing set, Rt was 0.79, RMSE t was 0.19%, and RPD was 1.61. This practice provides an efficient, low-cost approach with potentially high accuracy to estimate SOC contents and hence supports better management and protection strategies for desert wetland ecosystems.
| S-EPMC6195798 | biostudies-literature
Project description:Nitrogen is one of the most important nutrient indicators for the growth of crops, and is closely related to the chlorophyll content of leaves and thus influences the photosynthetic ability of the crops. In this study, five hybrid rice varieties were cultivated during one entire growing period in one experimental field supplied with six nitrogen fertilizer levels. Visible and near infrared (vis/NIR) reflectance spectroscopy combined with multivariate analysis was used to identify hybrid rice varieties and nitrogen fertilizer levels, as well as to detect chlorophyll content associated with nitrogen levels. The support vector machine (SVM) algorithm was applied to identify five varieties of hybrid rice and six levels of nitrogen fertilizer. The results demonstrated that different varieties of hybrid rice for each nitrogen level can be well distinguished except for the highest nitrogen level, and no nitrogen level for each rice variety can be completely identified from the other five nitrogen levels. Further, 12 spectral indices combined with partial least square (PLS) analysis were applied for estimating chlorophyll content of rice leaves from plants subjected to different nitrogen levels, and a root mean square error of cross-validation (RMSECV) of 0.506, a coefficient of determination (R 2) of 97.8% and a ratio of performance to deviation (RPD) of 4.6 for all rice varieties indicated this as a preferable procedure. This study demonstrates that Vis/NIR spectroscopy can have a great potential for identification of rice varieties and evaluation of nitrogen fertilizer levels.
| S-EPMC6837182 | biostudies-literature
Project description:Chlorophyll meters are widely used to guide nitrogen (N) management by monitoring leaf N status in agricultural systems, but the effects of environmental factors and leaf characteristics on leaf N estimations are still unclear. In the present study, we estimated the relationships among SPAD readings, chlorophyll content and leaf N content per leaf area for seven species grown in multiple environments. There were similar relationships between SPAD readings and chlorophyll content per leaf area for the species groups, but the relationship between chlorophyll content and leaf N content per leaf area, and the relationship between SPAD readings and leaf N content per leaf area varied widely among the species groups. A significant impact of light-dependent chloroplast movement on SPAD readings was observed under low leaf N supplementation in both rice and soybean but not under high N supplementation. Furthermore, the allocation of leaf N to chlorophyll was strongly influenced by short-term changes in growth light. We demonstrate that the relationship between SPAD readings and leaf N content per leaf area is profoundly affected by environmental factors and leaf features of crop species, which should be accounted for when using a chlorophyll meter to guide N management in agricultural systems.
| S-EPMC4548214 | biostudies-other
Project description:Regulated heat dissipation under excessive light comprises a complexity of mechanisms, whereby the supramolecular light-harvesting pigment-protein complex (LHC) shifts state from light harvesting towards heat dissipation, quenching the excess of photo-induced excitation energy in a non-photochemical way. Based on whole-leaf spectroscopy measuring upward and downward spectral radiance fluxes, we studied spectrally contiguous (hyperspectral) transient time series of absorbance A(λ,t) and passively induced chlorophyll fluorescence F(λ,t) dynamics of intact leaves in the visible and near-infrared wavelengths (VIS-NIR, 400-800 nm) after sudden strong natural-like illumination exposure. Besides light avoidance mechanism, we observed on absorbance signatures, calculated from simultaneous reflectance R(λ,t) and transmittance T(λ,t) measurements as A(λ,t) = 1 - R(λ,t) - T(λ,t), major dynamic events with specific onsets and kinetical behaviour. A consistent well-known fast carotenoid absorbance feature (500-570 nm) appears within the first seconds to minutes, seen from both the reflected (backscattered) and transmitted (forward scattered) radiance differences. Simultaneous fast Chl features are observed, either as an increased or decreased scattering behaviour during quick light adjustment consistent with re-organizations of the membrane. The carotenoid absorbance feature shows up simultaneously with a major F decrease and corresponds to the xanthophyll conversion, as quick response to the proton gradient build-up. After xanthophyll conversion (t = 3 min), a kinetically slower but major and smooth absorbance increase was occasionally observed from the transmitted radiance measurements as wide peaks in the green (~ 550 nm) and the near-infrared (~ 750 nm) wavelengths, involving no further F quenching. Surprisingly, in relation to the response to high light, this broad and consistent VIS-NIR feature indicates a slowly induced absorbance increase with a sigmoid kinetical behaviour. In analogy to sub-leaf-level observations, we suggest that this mechanism can be explained by a structure-induced low-energy-shifted energy redistribution involving both Car and Chl. These findings might pave the way towards a further non-invasive spectral investigation of antenna conformations and their relations with energy quenching at the intact leaf level, which is, in combination with F measurements, of a high importance for assessing plant photosynthesis in vivo and in addition from remote observations.
| S-EPMC6874624 | biostudies-literature
Project description:Estimating plasticity of leaf silicon (Si) in response to abiotic and biotic factors underpins our comprehension of plant defences and stress resistance in natural and agroecosystems. However, how nitrogen (N) addition and intraspecific plant-plant interactions affect Si concentration remains unclear.We grew 19 durum wheat genotypes (Triticum turgidum ssp. durum) in pots, either alone or in intra- or intergenotypic cultures of two individuals, and with or without N. Above-ground biomass, plant height and leaf [Si] were quantified at the beginning of the flowering stage.Nitrogen addition decreased leaf [Si] for most genotypes, proportionally to the biomass increase. Si plasticity to plant-plant interactions varied significantly among genotypes, with both increases and decreases in leaf [Si] when mixed with a neighbour, regardless of the mixture type (intra-/intergenotype). Besides, increased leaf [Si] in response to plant-plant interactions was associated with increased plant height.Our results suggest the occurrence of both facilitation and competition for Si uptake from the rhizosphere in wheat mixtures. Future research should identify which leaf and root traits characterise facilitating neighbours for Si acquisition. We also show that Si could be involved in height gain in response to intraspecific competition, possibly for increasing light capture. This important finding opens up new research directions on Si and plant-plant interactions in both natural ecosystems and agroecosystems. More generally, our results stress the need to explore leaf Si plasticity in responses to both abiotic and biotic factors to understand plant stress resistance. Read the free Plain Language Summary for this article on the Journal blog.
| S-EPMC9804457 | biostudies-literature
Project description:To develop a stable estimation model and identify effective wavelengths that could explain the variations in leaf nitrogen (N) concentration with different N supplies, growing seasons, ecological locations, growth stages, and wheat cultivars. Four field experiments were performed during two consecutive years (2017-2019) at three sites (Yuanyang, Hebi, and Wenxian) in Henan, China. In situ canopy spectral reflectance data under the aforementioned N supply conditions were obtained over a range of 400-950 nm (visible and near-infrared region). On the basis of the canopy raw spectral reflectance data and their subsequent transformation by two different techniques, first-derivative reflectance (FDR) and continuum removal (CR), four multivariate regression methods were comparatively analyzed and used to develop predictive models for estimating leaf N concentration: multiple linear regression (MLR), principal component regression (PCR), partial least square (PLS), and support vector machine (SVM). Results showed that leaf N concentration and canopy reflectance significantly varied with the levels of N fertilization, and a good correlation was observed for all the spectral techniques. Seven wavelengths with relatively higher r values than the bands of the raw spectra centered at 508, 525, 572, 709, 780, 876, and 925 nm were specified using the FDR technique. Based on the full wavelengths, the FDR-SVM model exhibited a good performance for leaf N concentration estimation, with coefficients of determination (r 2 val) for the validation datasets and corresponding relative percent deviations (RPDval) values of 0.842 and 2.383, respectively. However, the FDR-PLS yielded a more accurate assessment of the leaf N concentration than did the other methods, with r 2 val and RPDval values of 0.857 and 2.535, respectively. The variable importance in projection (VIP) scores from the FDR-PLS with the all canopy spectral region were used to screen the effective wavelengths of the spectral data. Therefore, six effective wavelengths centered at 525, 573, 710, 780, 875, and 924 nm were identified for leaf N concentration estimation. The SVM regression method with the effective wavelengths showed excellent performance for leaf N concentration estimation with r 2 val = 0.823 and RPDval = 2.280. These results demonstrated that the in situ canopy spectral technique is promising for the estimation of leaf N concentration in winter wheat based on the FDR-PLS regression model and the effective wavelengths identified.
| S-EPMC7333249 | biostudies-literature
Project description:Soil salinization is one of the most common forms of land degradation. The detection and assessment of soil salinity is critical for the prevention of environmental deterioration especially in arid and semi-arid areas. This study introduced the fractional derivative in the pretreatment of visible and near infrared (VIS-NIR) spectroscopy. The soil samples (n = 400) collected from the Ebinur Lake Wetland, Xinjiang Uyghur Autonomous Region (XUAR), China, were used as the dataset. After measuring the spectral reflectance and salinity in the laboratory, the raw spectral reflectance was preprocessed by means of the absorbance and the fractional derivative order in the range of 0.0-2.0 order with an interval of 0.1. Two different modeling methods, namely, partial least squares regression (PLSR) and random forest (RF) with preprocessed reflectance were used for quantifying soil salinity. The results showed that more spectral characteristics were refined for the spectrum reflectance treated via fractional derivative. The validation accuracies showed that RF models performed better than those of PLSR. The most effective model was established based on RF with the 1.5 order derivative of absorbance with the optimal values of R2 (0.93), RMSE (4.57 dS m-1), and RPD (2.78 ≥ 2.50). The developed RF model was stable and accurate in the application of spectral reflectance for determining the soil salinity of the Ebinur Lake wetland. The pretreatment of fractional derivative could be useful for monitoring multiple soil parameters with higher accuracy, which could effectively help to analyze the soil salinity.
| S-EPMC5936634 | biostudies-literature
Project description:Timely monitoring nitrogen status of rice crops with remote sensing can help us optimize nitrogen fertilizer management and reduce environmental pollution. Recently, the use of near-surface imaging spectroscopy is emerging as a promising technology that can collect hyperspectral images with spatial resolutions ranging from millimeters to decimeters. The spatial resolution is crucial for the efficiency in the image sampling across rice plants and the separation of leaf signals from the background. However, the optimal spatial resolution of such images for monitoring the leaf nitrogen concentration (LNC) in rice crops remains unclear. To assess the impact of spatial resolution on the estimation of rice LNC, we collected ground-based hyperspectral images throughout the entire growing season over 2 consecutive years and generated ten sets of images with spatial resolutions ranging from 1.3 to 450 mm. These images were used to determine the sensitivity of LNC prediction to spatial resolution with three groups of vegetation indices (VIs) and two multivariate methods Gaussian Process regression (GPR) and Partial least squares regression (PLSR). The reflectance spectra of sunlit-, shaded-, and all-leaf leaf pixels separated from background pixels at each spatial resolution were used to predict LNC with VIs, GPR and PLSR, respectively. The results demonstrated all-leaf pixels generally exhibited more stable performance than sunlit- and shaded-leaf pixels regardless of estimation approaches. The predictions of LNC required stage-specific LNC~VI models for each vegetative stage but could be performed with a single model for all the reproductive stages. Specifically, most VIs achieved stable performances from all the resolutions finer than 14 mm for the early tillering stage but from all the resolutions finer than 56 mm for the other stages. In contrast, the global models for the prediction of LNC across the entire growing season were successfully established with the approaches of GPR or PLSR. In particular, GPR generally exhibited the best prediction of LNC with the optimal spatial resolution being found at 28 mm. These findings represent significant advances in the application of ground-based imaging spectroscopy as a promising approach to crop monitoring and understanding the effects of spatial resolution on the estimation of rice LNC.
| S-EPMC6041568 | biostudies-literature
Project description:Leaf nitrogen concentration (LNC) is a major indicator in the estimation of the crop growth status which has been diffusely applied in remote sensing. Thus, it is important to accurately obtain LNC by using passive or active technology. Laser-induced fluorescence can be applied to monitor LNC in crops through analysing the changing of fluorescence spectral information. Thus, the performance of fluorescence spectrum (FS) and first-derivative fluorescence spectrum (FDFS) for paddy rice (Yangliangyou 6 and Manly Indica) LNC estimation was discussed, and then the proposed FS + FDFS was used to monitor LNC by multivariate analysis. The results showed that the difference between FS (R 2 = 0.781, s.d. = 0.078) and FDFS (R 2 = 0.779, s.d. = 0.097) for LNC estimation by using the artificial neural network is not obvious. The proposed FS + FDFS can improved the accuracy of LNC estimation to some extent (R 2 = 0.813, s.d. = 0.051). Then, principal component analysis was used in FS and FDFS, and extracted the main fluorescence characteristics. The results indicated that the proposed FS + FDFS exhibited higher robustness and stability for LNC estimation (R 2 = 0.851, s.d. = 0.032) than that only using FS (R 2 = 0.815, s.d. = 0.059) or FDFS (R 2 = 0.801, s.d. = 0.065).
| S-EPMC7062071 | biostudies-literature
Project description:While host gene expression has been related to symptoms associated with different phytoplasma diseases, it is unknown why some phytoplasmas are associated with different symptoms in genotypes of the same plant species. Pear tree selections showed symptoms of either leaf reddening (selection 8824-1) or leaf curling (selection 9328-1) associated with pear decline (PD) phytoplasma presence. PD populations were similar in leaves and shoots of the two selections, but in the roots, populations were significantly lower in selection 8824-1 than in 9328-1, indicating greater resistance. For host carbohydrate metabolism gene expression in PD-infected tissues, significant up-regulation in selection 8824-1 was observed for a sucrose synthase gene in leaves and an acid invertase gene in leaves and roots. These features have been associated with localized higher sugar levels in phytoplasma-infected tissues, and thus may be related to leaf reddening. For host stress/defense response gene expression in PD-infected tissues, significant up-regulation of a phenylalanine ammonia lyase gene was observed in PD-infected shoots of both selections; however, up-regulation of alcohol dehydrogenase gene in shoots, a chitinase gene in all tissues and a phloem protein 2 gene in roots was only observed for selection 8824-1. These changes indicate greater triggered innate immunity in roots associated with lower PD populations and leaf reddening. Leaf reddening may be related to changes in gene expression associated with increased sugar levels in leaves and stronger immune responses in several tissues, while leaf curling may be due to water stress resulting from dysfunctional root associated with higher PD populations in the roots.
| S-EPMC8838474 | biostudies-literature