Tiezhu Shi

Estimation of arsenic in agricultural soils using hyperspectral vegetation indices of rice

This study systematically analyzed the performance of multivariate hyperspectral vegetation indices of rice (Oryza sativa L.) in estimating the arsenic content in agricultural soils. Field canopy reflectance spectra was obtained in the jointing-booting growth stage of rice. Newly developed and published multivariate vegetation indices were initially calculated to estimate soil arsenic content. The well-performing vegetation indices were then selected using successive projections algorithm (SPA), and the SPA selected vegetation indices were adopted to calibrate a multiple linear regression model for estimating soil arsenic content. Results showed that a three-band vegetation index (R716-R568)/(R552-R568) performed best in the newly developed vegetation indices in estimating soil arsenic content. The photochemical reflectance index (PRI) and red edge position (REP) performed well in the published vegetation indices. Moreover, the linear combination of two vegetation indices ((R716-R568)/(R552-R568) and REP) selected using SPA improved the estimation of soil arsenic content. These results indicated that the newly developed three-band vegetation index (R716-R568)/(R552-R568) might be recommended as an indicator for estimating soil arsenic content in the study area. PRI and REP could be used as universal vegetation indices for monitoring soil arsenic contamination.

Evaluating different methods for grass nutrient estimation from canopy hyperspectral reflectance

The characterization of plant nutrients is important to understand the process of plant growth in natural ecosystems. This study attempted to evaluate the performances of univariate linear regression with various vegetation indices (VIs) and multivariate regression methods in estimating grass nutrients (i.e., nitrogen (N) and phosphorus (P)) with canopy hyperspectral reflectance. Synthetically considering predictive accuracy, simplicity, robustness and interpretation, the successive projections algorithm coupled with multiple linear regression (SPA-MLR) method was considered optimal for grass nutrient estimation at the canopy level, when compared with the performances of 12 statistical modeling methods, i.e., univariate linear regression with nine published VIs and three classical multivariate regression methods (stepwise multiple linear regression (SMLR), partial least squares regression (PLSR) and support vector regression (SVR)). The simple ratio index ( , is derivative reflectance) model had comparable performance to SPA-MLR model for P estimation. SPA-MLR provided comparable prediction accuracies with only three first derivative spectral bands for N (715, 731 and 2283 nm) and P (714, 729 and 1319 nm) estimations, compared with PLSR and SVR models, which used the full spectrum. Moreover, SPA-MLR provided robust prediction with the lowest bias values for N (−0.007%) and P (0.001%) estimations, and the fitting line between predicted and measured values was closer to the 1:1 line than the other models. Finally, most of the bands selected by SPA-MLR indirectly relate to foliar chlorophyll content, which suggests good physical interpretation.

A Wavelet-Based Area Parameter for Indirectly Estimating Copper Concentration in Carex Leaves from Canopy Reflectance

Due to the absence of evident absorption features and low concentrations, the copper (Cu) concentration in plant leaves has rarely been estimated from hyperspectral remote sensing data. The capability of remotely-sensed estimation of foliar Cu concentrations largely depends on its close relation to foliar chlorophyll concentration. To enhance the subtle spectral changes related to chlorophyll concentration under Cu stress, this study described a wavelet-based area parameter (SWT (605−720), the sum of reconstructed detail reflectance at fourth decomposition level over 605−720 nm using discrete wavelet transform) from the canopy hyperspectral reflectance (350−2500 nm, N = 71) of Carex (C. cinerascens). The results showed that Cu concentrations had negative and strong correlation with chlorophyll concentrations (r = －0.719, p < 0.001). Based on 1000 random dataset partitioning experiments, the 1000 linear calibration models provided a mean R2Val (determination coefficient of validation) value of 0.706 and an RPD (residual prediction deviation) value of 1.75 for Cu estimation. The bootstrapping and ANOVA test results showed that SWT (605−720) significantly (p < 0.05) outperformed published chlorophyll-related and wavelet-based spectral parameters. It was concluded here that the wavelet-based area parameter (i.e., SWT (605−720)) has potential ability to indirectly estimate Cu concentrations in Carex leaves through the strong correlation between Cu and chlorophyll. The method presented in this pilot study may be used to estimate the concentrations of other heavy metals. However, further research is needed to test its transferability and robustness for estimating Cu concentrations on other plant species in different biological and environmental conditions.

Application of Sentinel 2 MSI Images to Retrieve Suspended Particulate Matter Concentrations in Poyang Lake

Suspended particulate matter (SPM) is one of the dominant water constituents in inland and coastal waters, and SPM concnetration (CSPM) is a key parameter describing water quality. This study, using in-situ spectral and CSPM measurements as well as Sentinel 2 Multispectral Imager (MSI) images, aimed to develop CSPM retrieval models and further to estimate the CSPM values of Poyang Lake, China. Sixty-eight in-situ hyperspectral measurements and relative spectral response function were applied to simulate Sentinel 2 MIS spectra. Thirty-four samples were used to calibrate and the left samples were used to validate CSPM retrieval models, respectively. The developed models were then applied to two Sentinel 2 MSI images captured in wet and dry seasons, and the derived CSPM values were compared with those derived from MODIS B1 (λ = 645 nm). Results showed that the Sentinel 2 MSI B4–B8b models achieved acceptable to high fitting accuracies, which explained 81–93% of the variation of CSPM. The validation results also showed the reliability of these six models, and the estimated CSPM explained 77–93% of the variation of measured CSPM with the mean absolute percentage error (MAPE) ranging from 36.87% to 21.54%. Among those, a model based on B7 (λ = 783 nm) appeared to be the most accurate one. The Sentinel 2 MSI-derived CSPM values were generally consistent in spatial distribution and magnitude with those derived from MODIS. The CSPM derived from Sentinel 2 MSI B7 showed the highest consistency with MODIS on 15 August 2016, while the Sentinel 2 MSI B4 (λ = 665 nm) produced the highest consistency with MODIS on 2 April 2017. Overall, this study demonstrated the applicability of Sentinel 2 MSI for CSPM retrieval in Poyang Lake, and the Sentinel 2 MSI B4 and B7 are recommended for low and high loadings of SPM, respectively.

Improving Spectral Estimation of Soil Organic Carbon Content through Semi-Supervised Regression

Visible and near infrared (VIS-NIR) spectroscopy has been applied to estimate soil organic carbon (SOC) content with many modeling strategies and techniques, in which a crucial and challenging problem is to obtain accurate estimations using a limited number of samples with reference values (labeled samples). To solve such a challenging problem, this study, with Honghu City (Hubei Province, China) as a study area, aimed to apply semi-supervised regression (SSR) to estimate SOC contents from VIS-NIR spectroscopy. A total of 252 soil samples were collected in four field campaigns for laboratory-based SOC content determinations and spectral measurements. Semi-supervised regression with co-training based on least squares support vector machine regression (Co-LSSVMR) was applied for spectral estimations of SOC contents, and it was further compared with LSSVMR. Results showed that Co-LSSVMR could improve the estimations of SOC contents by exploiting samples without reference values (unlabeled samples) when the number of labeled samples was not excessively small and produce better estimations than LSSVMR. Therefore, SSR could reduce the number of labeled samples required in calibration given an accuracy threshold, and it holds advantages in SOC estimations from VIS-NIR spectroscopy with a limited number of labeled samples. Considering the increasing popularity of airborne platforms and sensors, SSR might be a promising modeling technique for SOC estimations from remotely sensed hyperspectral images.

Proximal and remote sensing techniques for mapping of soil contamination with heavy metals

ABSTRACT Heavy metal soil contamination is a severe environmental problem globally, and its mapping is vital for environmental managers and policymakers to determine its distributions and hotspots. This paper reviewed multiple proximal and remote sensing spectroscopy for convenient and inexpensive method of obtaining soil reflectance spectroscopy or environmental covariates, which can be used for mapping heavy metal soil contamination. Furthermore, spatial prediction using proximal remote-sensed data and environmental covariates was discussed. We suggested that mapping of the spatial distributions of metal species may be important due to the different bioavailabilities and toxicities of various species. The assimilation of multiple proximal/remote-sensed sensors may promote the horizontal and vertical mapping of soil heavy metals. Moreover, combining the advantages of satellite and unmanned aerial vehicle-based hyperspectral imaging systems will facilitate the development of a space–aeronautic incorporation hyperspectral observation technology that can monitor soil environment rapidly and accurately at a large scale.

Improving Land Use/Land Cover Classification by Integrating Pixel Unmixing and Decision Tree Methods

Decision tree classification is one of the most efficient methods for obtaining land use/land cover (LULC) information from remotely sensed imageries. However, traditional decision tree classification methods cannot effectively eliminate the influence of mixed pixels. This study aimed to integrate pixel unmixing and decision tree to improve LULC classification by removing mixed pixel influence. The abundance and minimum noise fraction (MNF) results that were obtained from mixed pixel decomposition were added to decision tree multi-features using a three-dimensional (3D) Terrain model, which was created using an image fusion digital elevation model (DEM), to select training samples (ROIs), and improve ROI separability. A Landsat-8 OLI image of the Yunlong Reservoir Basin in Kunming was used to test this proposed method. Study results showed that the Kappa coefficient and the overall accuracy of integrated pixel unmixing and decision tree method increased by 0.093% and 10%, respectively, as compared with the original decision tree method. This proposed method could effectively eliminate the influence of mixed pixels and improve the accuracy in complex LULC classifications.

Geo-detection of factors controlling spatial patterns of heavy metals in urban topsoil using multi-source data

Heavy metal contamination has become a serious and widespread problem in urban environment. Understanding its controlling factors is vital for the identification, prevention, and remediation of pollution sources. This study aimed to identify the factors controlling heavy metal accumulation in urban topsoil using the geodetector method and multiple data sources. Environmental factors including geology, relief (elevation, slope, and aspect), and organism (land-use and vegetation) were extracted from a geological thematic map, digital elevation model, and time-series of Landsat images, respectively. Then, the power of determinant (q) was calculated using geodetector to measure the affinity between the environmental factors and arsenic (As) and lead (Pb). Geology was the dominant factor for As distribution in the this study area; it explained 38% of the spatial variation in As, and nonlinear enhancements were observed for the interactions between geology and elevation (q = 0.50) and slope (q = 0.49). Land-use and vegetation bi-enhanced each other and explained 39% of the spatial variation in Pb. These results indicated that geology and relief were the factors controlling the spatial distribution of As, and organism factors, especially anthropogenic activities, were the factors controlling the spatial distribution of Pb in the study area. As was derived from weathering transportation, and deposition processes of original bedrock and subsequent pedogenesis, and anthropogenic activity was the most likely source of Pb contamination in urban topsoil in Shenzhen. Moreover, geodetector provided evidence to explore the factors controlling spatial patterns of heavy metals in soils.

Comparison of Machine Learning Techniques in Inferring Phytoplankton Size Classes

The size of phytoplankton not only influences its physiology, metabolic rates and marine food web, but also serves as an indicator of phytoplankton functional roles in ecological and biogeochemical processes. Therefore, some algorithms have been developed to infer the synoptic distribution of phytoplankton cell size, denoted as phytoplankton size classes (PSCs), in surface ocean waters, by the means of remotely sensed variables. This study, using the NASA bio-Optical Marine Algorithm Data set (NOMAD) high performance liquid chromatography (HPLC) database, and satellite match-ups, aimed to compare the effectiveness of modeling techniques, including partial least square (PLS), artificial neural networks (ANN), support vector machine (SVM) and random forests (RF), and feature selection techniques, including genetic algorithm (GA), successive projection algorithm (SPA) and recursive feature elimination based on support vector machine (SVM-RFE), for inferring PSCs from remote sensing data. Results showed that: (1) SVM-RFE worked better in selecting sensitive features; (2) RF performed better than PLS, ANN and SVM in calibrating PSCs retrieval models; (3) machine learning techniques produced better performance than the chlorophyll-a based three-component method; (4) sea surface temperature, wind stress, and spectral curvature derived from the remote sensing reflectance at 490, 510, and 555 nm were among the most sensitive features to PSCs; and (5) the combination of SVM-RFE feature selection techniques and random forests regression was recommended for inferring PSCs. This study demonstrated the effectiveness of machine learning techniques in selecting sensitive features and calibrating models for PSCs estimations with remote sensing.

Spectroscopic Diagnosis of Arsenic Contamination in Agricultural Soils

This study investigated the abilities of pre-processing, feature selection and machine-learning methods for the spectroscopic diagnosis of soil arsenic contamination. The spectral data were pre-processed by using Savitzky-Golay smoothing, first and second derivatives, multiplicative scatter correction, standard normal variate, and mean centering. Principle component analysis (PCA) and the RELIEF algorithm were used to extract spectral features. Machine-learning methods, including random forests (RF), artificial neural network (ANN), radial basis function- and linear function- based support vector machine (RBF- and LF-SVM) were employed for establishing diagnosis models. The model accuracies were evaluated and compared by using overall accuracies (OAs). The statistical significance of the difference between models was evaluated by using McNemar’s test (Z value). The results showed that the OAs varied with the different combinations of pre-processing, feature selection, and classification methods. Feature selection methods could improve the modeling efficiencies and diagnosis accuracies, and RELIEF often outperformed PCA. The optimal models established by RF (OA = 86%), ANN (OA = 89%), RBF- (OA = 89%) and LF-SVM (OA = 87%) had no statistical difference in diagnosis accuracies (Z < 1.96, p < 0.05). These results indicated that it was feasible to diagnose soil arsenic contamination using reflectance spectroscopy. The appropriate combination of multivariate methods was important to improve diagnosis accuracies.