Handheld Spectroradiometer Research Articles

Chlorophyll estimation using multi-spectral unmanned aerial system based on machine learning techniques

Remote sensing for precision agriculture is a proven tool for efficient management of crop inputs. High spatial and temporal resolutions are requisite for accurate and timely estimate of crop parameters. We made an attempt to estimatetheleaf chlorophyll concentration of standing maize plant from high resolution (5 cm) multi-spectral Unmanned Aerial Vehicle (UAV) images. The UAV images in Green, Red, Red-Edge and NIR bands of standing maize fields under various nutrient induced abiotic stresses were acquired by flying a hexacopter at an elevation of 60 m at ICAR Research Complex for NEH Region, Meghalaya. A handheld spectroradiometer was also used to record the spectra in 1024 bands, ranging from 350 nm to 2500 nm to support the UAV study. We evaluated advanced machine learning algorithms combined with spectral data and ground truth chlorophyll to model the chlorophyll estimates. Algorithms included Support vector regression, Relevance vector regression, Gaussian process regression, Kernel ridge regression and Random forest with K-fold cross validation. The multivariate analysis applied on spectroradiometer and UAV data showed the dominance of Red-band for chlorophyll prediction with R2 values greater than 0.80. Among the machine learning algorithms, we found the Kernel-Ridge regression was most robust method for developing chlorophyll estimation model with minimal RMSE (0.057 mg/gm) and regression coefficient of determination (R2 = 0.904). The relevance vector machine also predicted chlorophyll concentration satisfactorily (R2 = 0.87 with RMSE of 0.06 mg/gm), but took larger training time. The optimization and hybridisation of kernel based algorithms is further needed to enhance the reliability of models for prediction of leaf chlorophyll concentrations.

Dual Activation Function-Based Extreme Learning Machine (ELM) for Estimating Grapevine Berry Yield and Quality

Reliable assessment of grapevine productivity is a destructive and time-consuming process. In addition, the mixed effects of grapevine water status and scion-rootstock interactions on grapevine productivity are not always linear. Despite the potential opportunity of applying remote sensing and machine learning techniques to predict plant traits, there are still limitations to previously studied techniques for vine productivity due to the complexity of the system not being adequately modeled. During the 2014 and 2015 growing seasons, hyperspectral reflectance spectra were collected using a handheld spectroradiometer in a vineyard designed to investigate the effects of irrigation level (0%, 50%, and 100%) and rootstocks (1103 Paulsen, 3309 Couderc, SO4 and Chambourcin) on vine productivity. To assess vine productivity, it is necessary to measure factors related to fruit ripeness and not just yield, as an over cropped vine may produce high-yield but poor-quality fruit. Therefore, yield, Total Soluble Solids (TSS), Titratable Acidity (TA) and the ratio TSS/TA (maturation index, IMAD) were measured. A total of 20 vegetation indices were calculated from hyperspectral data and used as input for predictive model calibration. Prediction performance of linear/nonlinear multiple regression methods and Weighted Regularized Extreme Learning Machine (WRELM) were compared with our newly developed WRELM-TanhRe. The developed method is based on two activation functions: hyperbolic tangent (Tanh) and rectified linear unit (ReLU). The results revealed that WRELM and WRELM-TanhRe outperformed the widely used multiple regression methods when model performance was tested with an independent validation dataset. WRELM-TanhRe produced the highest prediction accuracy for all the berry yield and quality parameters (R2 of 0.522–0.682 and RMSE of 2–15%), except for TA, which was predicted best with WRELM (R2 of 0.545 and RMSE of 6%). The results demonstrate the value of combining hyperspectral remote sensing and machine learning methods for improving of berry yield and quality prediction.

Low-cost assessment of grain yield in durum wheat using RGB images

The pattern of photosynthetic area of the canopy throughout the crop cycle is an important factor for determining grain yield in wheat. This work proposes the use of zenithal RGB images of the canopy taken in natural light conditions to derive vegetation indices as a low-cost approach to predict grain yield. A set of 23 varieties of durum wheat was monitored in three growing conditions (support irrigation, rainfed and late planting) and two sites (Aranjuez and Valladolid, Spain), totalling 6 field trials. For each plot, digital RGB images were taken periodically from seedling emergence to late grain filling. RGB-based Green Area (GA), Greener Area (GGA), Normalized Green Red Difference Index (NGRDI), Triangular Greenness Index (TGI) and a novel photosynthetic area index based on the CIE L*u*v* colour space (u*v*A) were compared to handheld spectroradiometer Normalised Difference Vegetation Index (NDVI) for reference. In the case of the irrigated and late planting trials the best phenotypic predictions of grain yield were achieved with the vegetation indices measured during the last part of the crop cycle (i.e. grain filling). For the rainfed trials the best phenotypic predictions were achieved with indices measured earlier (around heading). Among all the evaluated indices, the novel index performed the best. Considering the heritabilities of the evaluated RGB indices and their genetic correlations with grain yield, index-based predictions of grain yield were best in the early crop stages for both rainfed and irrigated conditions, while for late planting indices measured at different crop stages performed equally well.

Estimating chlorophyll content at leaf scale in viroid-inoculated oil palm seedlings (Elaeis guineensis Jacq.) using reflectance spectra (400 nm–1050 nm)

ABSTRACTOrange Spotting (OS) is an emerging disease in oil palm (Elaeis guineensis Jacq) that has been associated with Coconut cadang-cadang viroid (CCCVd; Cocadviroid, Pospiviroidae). This study was aimed at selecting efficient wavebands in the Visible/near infrared (VNIR) spectrum (400–1050 nm) for estimation of chlorophyll content in CCCVd-inoculated oil palm seedlings. The VNIR spectrum is most sensitive to chlorophyll stress and able to differentiate between healthy and diseased plants. Therefore, a greenhouse experiment was conducted to remotely sense the chlorophyll content using a hyperspectral hand-held spectroradiometer (Analytical Spectral Device (ASD) FieldSpec–II) and a portable chlorophyll meter (Minolta Soil-Plant Analysis Development (SPAD)−502). FieldSpec–II provides reflectance spectra with narrow and contiguous wavebands while SPAD-502 churns out SPAD readings corresponding to chlorophyll content. Interval Partial Least Squares (iPLS) model was used to assess the relationship between independent variable (spectra) and dependent variable (SPAD readings). Five different spectral datasets (i.e. raw spectra, first-order derivatives, Savitzky-Golay smoothing, Multiplicative Scatter Correction (MSC), and standard normal transformation) were investigated. Results showed that the MSC dataset yielded the most accurate estimation with a root mean square error of prediction of 3.70% and a correlation coefficient for prediction of 0.72. A total of thirty wavebands within a range between 601 nm and 630 nm was selected. This work showed that VNIR spectra has the potential to diagnose OS disease in oil palm using the iPLS regression method. This is completely a new application of hyperspectral remote sensing for the oil palm industry, particularly in the domain of plant protection.

Mapping seaweed beds using multispectral imagery retrieved by unmanned aerial vehicles

Mapping seaweed beds using multispectral imagery retrieved by unmanned aerial vehicles

Determining soil organic carbon concentration in agricultural fields using a handheld spectroradiometer: Implication for soil fertility measurement

The soil organic carbon (SOC) plays a vital role in plant growth and development, and therefore is considered as one of the most important indicators of soil quality. This study was carried out in the central region of Saudi Arabia to explore the potential of spectroscopy in determining the SOC concentration in low-fertility soils. Spectral reflectance data were collected, under the controlled laboratory conditions on 39 air-dried 2.0 mm sieved soil samples, using a handheld spectroradiometer of a wavelength range between 350 nm and 2500 nm in the direct contact probe mode. The concentration of the SOC was determined using the Walkley and Black (W&B) and the UV-VIS spectrophotometric methods. The increase in the concentration of SOC was associated with a decrease in the corresponding spectral reflectance. Regression analysis showed linear relationships with high significant correlation between the spectral reflectance and the SOC measured by both the UV-VIS (Model-1: R2=0.46, p=0.00015 and RMSE=6.6 g/kg) and the W&B (Model-2: R2=0.48, p=8.92E-05 and RMSE= 2.8 g/kg) methods. For these models, two wavebands with wavelengths of 2167 nm (Model-1) and 1359 nm (Model-2) were identified as the most sensitive bands to the SOC concentration. The cross-validation confirmed the validity of Model-1 with R2, p and RMSE values of 0.50, 0.0099 and 6.6 g/kg, respectively. The validation results of the Model-2 showed values of R2, p and RMSE of 0.72, 0.00023 and 4.0 g/kg, respectively. Results of this study revealed the possibility and the potential of using the spectral reflectance technique in predicting the concentration of SOC. Keywords: soil fertility, organic carbon, modeling, spectroscopy, reflectance DOI: 10.25165/j.ijabe.20181106.4061 Citation: Tola E, Al-Gaadi K A, Madugundu R, Kayad A G, Alameen A A, Edrees H F, et al. Determining soil organic carbon concentration in agricultural fields using a handheld spectroradiometer: Implication for soil fertility measurement. Int J Agric & Biol Eng, 2018; 11(6): 13–19.

Retrieval of Maize Leaf Area Index Using Hyperspectral and Multispectral Data

Field spectra acquired from a handheld spectroradiometer and Sentinel-2 images spectra were used to investigate the applicability of hyperspectral and multispectral data in retrieving the maize leaf area index in low-input crop systems, with high spatial and intra-annual variability, and low yield, in southern Mozambique, during three years. Seventeen vegetation indices, comprising two and three band indices, and nine machine learning regression algorithms (MLRA) were tested for the statistical approach while five cost functions were tested in the look-up-table (LUT) inversion approach. The three band vegetation indices were selected, specifically the modified difference index (mDId: 725; 715; 565) for the hyperspectral dataset and the modified simple ratio (mSRc: 740; 705; 865) for the multispectral dataset of field spectra and the three band spectral index (TBSIb: 665; 865; 783) for the Sentinel-2 dataset. The relevant vector machine was the selected MLRA for the two datasets of field spectra (multispectral and hyperspectral) while the support vector machine was selected for the Sentinel-2 data. When using the LUT inversion technique, the minimum contrast estimation and the Bhattacharyya divergence cost functions were the best performing. The vegetation indices outperformed the other two approaches, with the TBSIb as the most accurate index (RMSE = 0.35). At the field scale, spectral data from Sentinel-2 can accurately retrieve the maize leaf area index in the study area.

Determination of rubber-tree clones leaf diseases spectral using Unmanned Aerial Vehicle compact sensor

Currently, one of the remote sensing platform that adequatly been used is Unmanned Aerial Vehicle (UAV) which suitable in monitoring and mapping for agriculture sector at large area payload by compact sensor. Thus, this study is deploy the UAV compact sensor to identify the characteristics of rubber tree clone leaf diseases based on two groups of spectral wavelength which is visible (RGB: 0.4 µm – 0.7 µm) and near infrared (NIR: 0.7µm – 2.0 µm), respectively. Spectral obtained using UAV platform is then to be validated with ground observation handheld spectroradiometer. Eight types of rubber tree clones leaf at three different conditions (healty, unhealty and severe) were randomly selected within the 9.4 hectare Experimental Rubber Plot, Rubber Research Institute of Malaysia (RRIM), Kota Tinggi, Johor whereby consist RRIM 2000 series, RRIM 3000 series and PB series, respectively. As a result, this study has found that the spectral trend based on UAV compact sensor from eight different types of rubber tree clones leaf shows the similarity to the general basic vegetation spectral in transition from blue to NIR spectral. However, referring to spectral obtained from handheld spectroradiometer there is no drastically changes of spectral in visible region but drastically increase in NIR region. Thus, this study has conclude that the spectral signature characteristics for healthy, unhealthy and severe for leaf diseases from every single rubber tree clones can be identified obviously in NIR region using UAV compact sensor.

Detection of Firmiana danxiaensis Canopies by a Customized Imaging System Mounted on an UAV Platform

The objective of this study was to test the effectiveness of mapping the canopies of Firmiana danxiaensis (FD), a rare and endangered plant species in China, from remotely sensed images acquired by a customized imaging system mounted on an unmanned aerial vehicle (UAV). The work was conducted in an experiment site (approximately 10 km2) at the foot of Danxia Mountain in Guangdong Province, China. The study was conducted as an experimental task for a to-be-launched large-scale FD surveying on Danxia Mountain (about 200 km2 in area) by remote sensing on UAV platforms. First, field-based spectra were collected through hand-held hyperspectral spectroradiometer and then analyzed to help design a classification schema which was capable of differentiating the targeted plant species in the study site. Second, remote-sensed images for the experiment site were acquired and calibrated through a variety of preprocessing steps. Orthoimages and a digital surface model (DSM) were generated as input data from the calibrated UAV images. The spectra and geometry features were used to segment the preprocessed UAV imagery into homogeneous patches. Lastly, a hierarchical classification, combined with a support vector machine (SVM), was proposed to identify FD canopies from the segmented patches. The effectiveness of the classification was evaluated by on-site GPS recordings. The result illustrated that the proposed hierarchical classification schema with a SVM classifier on the remote sensing imagery collected by the imaging system on UAV provided a promising method for mapping of the spatial distribution of the FD canopies, which serves as a replacement for field surveys in the attempt to realize a wide-scale plant survey by the local governments.

Determining soil organic carbon concentration in agricultural fields using a handheld spectroradiometer: Implication for soil fertility measurement

The soil organic carbon (SOC) plays a vital role in plant growth and development, and therefore is considered as one of the most important indicators of soil quality. This study was carried out in the central region of Saudi Arabia to explore the potential of spectroscopy in determining the SOC concentration in low-fertility soils. Spectral reflectance data were collected, under the controlled laboratory conditions on 39 air-dried 2.0 mm sieved soil samples, using a handheld spectroradiometer of a wavelength range between 350 nm and 2500 nm in the direct contact probe mode. The concentration of the SOC was determined using the Walkley and Black (W&B) and the UV-VIS spectrophotometric methods. The increase in the concentration of SOC was associated with a decrease in the corresponding spectral reflectance. Regression analysis showed linear relationships with high significant correlation between the spectral reflectance and the SOC measured by both the UV-VIS (Model-1: R2=0.46, p=0.00015 and RMSE=6.6 g/kg) and the W&B (Model-2: R2=0.48, p=8.92E-05 and RMSE= 2.8 g/kg) methods. For these models, two wavebands with wavelengths of 2167 nm (Model-1) and 1359 nm (Model-2) were identified as the most sensitive bands to the SOC concentration. The cross-validation confirmed the validity of Model-1 with R2, p and RMSE values of 0.50, 0.0099 and 6.6 g/kg, respectively. The validation results of the Model-2 showed values of R2, p and RMSE of 0.72, 0.00023 and 4.0 g/kg, respectively. Results of this study revealed the possibility and the potential of using the spectral reflectance technique in predicting the concentration of SOC. Keywords: soil fertility, organic carbon, modeling, spectroscopy, reflectance DOI: 10.25165/j.ijabe.20181106.4061 Citation: Tola E, Al-Gaadi K A, Madugundu R, Kayad A G, Alameen A A, Edrees H F, et al. Determining soil organic carbon concentration in agricultural fields using a handheld spectroradiometer: Implication for soil fertility measurement. Int J Agric & Biol Eng, 2018; 11(6): 13–19.

Spatial and temporal variations of the surface albedo and other factors influencing Urumqi Glacier No. 1 in Tien Shan, China

ABSTRACTGlacier albedo controls the surface energy budget, the variability of which affects the glacier surface melt rate and, in turn, impacts the mass balance of the glacier. During 2013 and 2014, spatial and temporal variations of albedo were investigated using 18 Landsat images of Urumqi Glacier No. 1. Factors influencing these spatiotemporal profiles were analyzed. An established retrieval process, including geolocation, radiometric calibration, atmospheric, topographic, and anisotropic correction and narrow- to broadband conversion, was applied for the first time to Landsat-8 images. Differences between Landsat image derived albedo values and albedo values measured using a handheld spectroradiometer ranged from −0.024 to 0.049. Spatial and temporal variations of surface albedo were significant, especially in the ablation area. The variability of the values of ice albedo ranged from 0.06 to 0.44 due to topographic effects and light-absorbing impurities. The results suggest that this retrieval method can be used to investigate the spatial and temporal variability of surface albedo from Landsat-8 images on mountain glaciers. Moreover, as constant albedo values for ice and snow cannot be assumed, the distribution of albedo was not completely dependent on altitude under conditions of more intense ablation, and by reason of light-absorbing impurities during the melt season.

Machine learning prediction of coffee rust severity on leaves using spectroradiometer data

The interest in using remote sensing data in agriculture, including plant disease assessments, has increased considerably in the last years. The satellite-based Sentinel-2 MultiSpectral Imager (MSI) sensor has been launched recently for multispectral vegetation condition assessment for agricultural and ecosystem applications. The aim of this pilot study conducted in the greenhouse using a hand-held spectroradiometer was to assess the utility of the same wavebands as used in the Sentinel-2 MSI in assessing and modeling coffee leaf rust (CLR) based on the non-linear radial basis function-partial least squares regression (RBF-PLS) machine learning algorithm, compared with ordinary partial least squared regression (PLSR). The RBF-PLS derived models satisfactorily described CLR severity (R 2=0.92 and RMSE=6.1% with all bands and R 2=0.78 and RMSE=10.2% with selected bands) when compared with PLSR (R 2 = 0.27 and RMSE = 18.7% with all bands and R 2 = 0.17 and RMSE = 19.8% with selected bands). Specifically, four bands, B2 (490 nm), B4 (665 nm), B5 (705 nm) and B7 (783 nm) were identified as the most important spectral bands in assessing and modeling CLR severity. Better accuracy was obtained for most severe levels of CLR (R 2=0.71 using all variables) than for moderate levels (R 2=0.38 using all variables). Overall, the findings of this study showed that the use of RBF-PLS and the four Sentinel-2 MSI bands could enhance CLR severity estimation at the leaf level. Further work will be needed to extrapolate these findings to the crop level using the Sentinel-2 platform.

An evaluation of a handheld spectroradiometer for the near real-time measurement of cyanobacteria for bloom management purposes.

A commercially available handheld spectroradiometer, the WISP-3, was assessed as a tool for monitoring freshwater cyanobacterial blooms for management purposes. Three small eutrophic urban ponds which displayed considerable within-pond and between-pond variability in water quality and cyanobacterial community composition were used as trial sites. On-board algorithms provide field measurements of phycocyanin (CPC) and chlorophyll-a (Chl-a) from surface reflectance spectra measured by the instrument. These were compared with laboratory measurements. Although significant but weak relationships were found between WISP-3 measured CPC and cyanobacterial biovolume measurements and WISP-3 and laboratory Chl-a measurements, there was considerable scatter in the data due likely to error in both WISP-3 and laboratory measurements. The relationships generally differed only slightly between ponds, indicating that different cyanobacterial communities had little effect on the pigment retrievals of the WISP-3. The on-board algorithms need appropriate modification for local conditions, posing a problem if it is to be used extensively across water bodies with differing optical properties. Although suffering a range of other limitations, the WISP-3 has a potential as a rapid screening tool for preliminary risk assessment of cyanobacterial blooms. However, such field assessment would still require adequate support by sampling and laboratory-based analysis.

Assessing maize foliar water stress levels under field conditions using in-situ spectroscopy

Plant physiological processes required for crop productivity are dependent on the availability of water to crops. Water availability to crops therefore requires real time monitoring for timeous rescue or intervention measures. Such monitoring over vast areas is only possible through remotely sensed techniques such as spectroscopy with its numerous fine wavelengths and is non-destructive to the crops as opposed to other traditional ground-based methods. The management of spectral reflectance data to extract information of importance for plant water status has been motivated by knowledge of the availability of specific bands in the electromagnetic spectrum responsible for water absorption. The purpose of this study was to investigate the potential of using selected spectral bands to develop water indices that could monitor the water status at leaf level on maize (Zea mays) plants grown under field conditions. Leaf spectral reflectance of maize plants was collected under three different water conditions being healthy (H), intermediary water stressed (IWS) and water stressed (WS) using a leaf-clip of a handheld spectroradiometer. The spectral reflectance indicated an increased reflectance in portions of the visible, near-infrared and short infrared regions of the electromagnetic spectrum for the water stressed maize plants. The random forest (RF) algorithm was utilised to extract wavelengths of importance from which water indices were developed among which were the normalised difference water index (NDWI860-1240) and the water band index (WBI950-970). The indices were used in a combined algorithm of RF and partial least square (PLS) for its predictive ability to classify the maize leaf water status into the three categories (H, IWS and WS). The results showed an overall accuracy of 70±1.2 %. Therefore, confirming the potential of assessing leaf water content using in-situ spectroscopy. The three most important indices were NDWI860-1240, NDWI860-1240 and NDWI860-1240. An in-depth study would be required to quantify and measure actual water content in maize leaves and possibly upscale to canopy level which would directly support irrigation management plans.Keywords: in-situ spectroscopy, maize, partial least square, random forest, water spectral indices, water stress

Hyperspectral-based predictive modelling of grapevine water status in the Portuguese Douro wine region

In this study, hyperspectral reflectance (HySR) data derived from a handheld spectroradiometer were used to assess the water status of three grapevine cultivars in two sub-regions of Douro wine region during two consecutive years. A large set of potential predictors derived from the HySR data were considered for modelling/predicting the predawn leaf water potential (Ψpd) through different statistical and machine learning techniques. Three HySR vegetation indices were selected as final predictors for the computation of the models and the in-season time trend was removed from data by using a time predictor. The vegetation indices selected were the Normalized Reflectance Index for the wavelengths 554nm and 561nm (NRI554;561), the water index (WI) for the wavelengths 900nm and 970nm, and the D1 index which is associated with the rate of reflectance increase in the wavelengths of 706nm and 730nm. These vegetation indices covered the green, red edge and the near infrared domains of the electromagnetic spectrum. A large set of state-of-the-art analysis and statistical and machine-learning modelling techniques were tested. Predictive modelling techniques based on generalized boosted model (GBM), bagged multivariate adaptive regression splines (B-MARS), generalized additive model (GAM), and Bayesian regularized neural networks (BRNN) showed the best performance for predicting Ψpd, with an average determination coefficient (R2) ranging between 0.78 and 0.80 and RMSE varying between 0.11 and 0.12MPa. When cultivar Touriga Nacional was used for training the models and the cultivars Touriga Franca and Tinta Barroca for testing (independent validation), the models performance was good, particularly for GBM (R2=0.85; RMSE=0.09MPa). Additionally, the comparison of Ψpd observed and predicted showed an equitable dispersion of data from the various cultivars. The results achieved show a good potential of these predictive models based on vegetation indices to support irrigation scheduling in vineyard.

Detecting the Early Stage of Phaeosphaeria Leaf Spot Infestations in Maize Crop Using In Situ Hyperspectral Data and Guided Regularized Random Forest Algorithm

Phaeosphaeria leaf spot (PLS) is considered one of the major diseases that threaten the stability of maize production in tropical and subtropical African regions. The objective of the present study was to investigate the use of hyperspectral data in detecting the early stage of PLS in tropical maize. Field data were collected from healthy and the early stage of PLS over two years (2013 and 2014) using a handheld spectroradiometer. An integration of a newly developed guided regularized random forest (GRRF) and a traditional random forest (RF) was used for feature selection and classification, respectively. The 2013 dataset was used to train the model, while the 2014 dataset was used as independent test dataset. Results showed that there were statistically significant differences in biochemical concentration between the healthy leaves and leaves that were at an early stage of PLS infestation. The newly developed GRRF was able to reduce the high dimensionality of hyperspectral data by selecting key wavelengths with less autocorrelation. These wavelengths are located at 420 nm, 795 nm, 779 nm, 1543 nm, 1747 nm, and 1010 nm. Using these variables (n=6), a random forest classifier was able to discriminate between the healthy maize and maize at an early stage of PLS infestation with an overall accuracy of 88% and a kappa value of 0.75. Overall, our study showed potential application of hyperspectral data, GRRF feature selection, and RF classifiers in detecting the early stage of PLS infestation in tropical maize.

Estimating Swiss chard foliar macro- and micronutrient concentrations under different irrigation water sources using ground-based hyperspectral data and four partial least squares (PLS)-based (PLS1, PLS2, SPLS1 and SPLS2) regression algorithms

Timely information on crop foliar nutrient content provides a measure of crop nutritional and vitality status. Growers and farm managers use such information for precision crop management such as an appropriate fertilizer application to correct for any crop nutrient deficiencies at identified hotspots. Foliar heavy nutrient content could also be a direct indicator of crops having been polluted from the surroundings, which may be a result of heavy metals absorbed from, among others, contaminated soils and waste water. In the present study, we explored the potential use of four partial least squares (PLS)-based regression algorithms for estimating foliar Swiss chard macro- and micronutrient concentrations using ground-based hyperspectral data under three treatments; i.e. rainwater+fertilizer (‘R+F’), tap water+fertilizer (‘T+F’), and treated wastewater (‘W’). Swiss chard canopy-level hyperspectral measurements under these three treatments were collected using a handheld spectroradiometer 2.5months after sowing. The reflectance spectra were normalized to their first-order derivatives. The concentrations of three Swiss chard foliar macronutrients (NPK) and three micronutrients (Zn, Cu and Fe) under the three treatments were determined. Regression models for estimating macro- and micronutrient concentrations were then derived using PLS1 and sparse PLS1 methods, while the potential simultaneous estimation of the macronutrient as well as micronutrient concentrations was explored using the PLS2 and SPLS2 regression approaches. Results showed that high variances in the macro- and micronutrient concentrations can be explained by the four regression models under the three treatments (R2train ranged between 0.73 and 0.99), except when P, Zn and Cu concentrations were estimated using the PLS2-based models under the three treatments (R2train ranged between 0.08 and 0.68) and Fe concentration using SPLSR1 under ‘W’ treatment (R2train=0.64). Our results further showed that Swiss chard foliar N (RMSE=1.67%) concentration under ‘R+F’ treatment and Fe (RMSE=7.83%) concentration under the ‘T+F’ treatment most accurately estimated macro- and micronutrients. Our study also showed that the Swiss chard foliar macronutrient concentrations were more accurately estimated compared to micronutrient concentrations and PLS2 outperformed the PLS1 based regression model. The results of the current study pave the way for developing an effective foliar nutrient estimation routine suitable for monitoring Swiss chard nutrient status under different treatments.

Detection and Differentiation between Laurel Wilt Disease, Phytophthora Disease, and Salinity Damage Using a Hyperspectral Sensing Technique

Laurel wilt (Lw) is a fatal disease. It is a vascular pathogen and is considered a major threat to the avocado industry in Florida. Many of the symptoms of Lw resemble those that are caused by other diseases or stress factors. In this study, the best wavelengths with which to discriminate plants affected by Lw from stress factors were determined and classified. Visible-near infrared (400–950 nm) spectral data from healthy trees and those with Lw, Phytophthora, or salinity damage were collected using a handheld spectroradiometer. The total number of wavelengths was averaged in two ranges: 10 nm and 40 nm. Three classification methods, stepwise discriminant (STEPDISC) analysis, multilayer perceptron (MLP), and radial basis function (RBF), were applied in the early stage of Lw infestation. The classification results obtained for MLP, with percent accuracy of classification as high as 98% were better than STEPDISC and RBF. The MLP neural network selected certain wavelengths that were crucial for correctly classifying healthy trees from those with stress trees. The results showed that there were sufficient spectral differences between laurel wilt, healthy trees, and trees that have other diseases; therefore, a remote sensing technique could diagnose Lw in the early stage of infestation.

Spectral considerations for modeling yield of canola

Prominent yellow flowers that are present in a Brassica oilseed crop such as canola require careful consideration when selecting a spectral index for yield estimation. This study evaluated spectral indices for multispectral sensors that correlate with the seed yield of Brassica oilseed crops. A small-plot experiment was conducted near Adams, Oregon in which spring canola was grown under varying water regimes and nitrogen treatments to create a wide range in oilseed yield. Plot measurements consisted of canopy reflectance at flowering using a hand-held spectroradiometer and seed yield at physiological maturity. Spectroradiometric measurements were converted to MODIS band equivalent reflectance. Selected indices were computed from spectra obtained with the radiometer and correlated with seed yield. A normalized difference yellowness index (NDYI), computed from the green and blue wavebands, overcame limitations of the normalized difference vegetation index (NDVI) during flowering and best modeled variability in relative yield potential. NDYI was more linear and correlated with county-wide oilseed yield data and MODIS satellite data from North Dakota (r2≤0.72) than NDVI (r2≤0.66). NDYI only requires wavebands in the visible region of the spectrum and can be applied to any satellite or aerial sensor that has blue and green channels. These findings highlight the benefit of using a spectral index that is sensitive to reproductive growth of vegetation instead of vegetative growth for crops with spectrally prominent reproductive canopy elements. Our results indicate that NDYI is a better indicator of yield potential than NDVI during mid-season development stages, especially peak flowering.

Inversion of radiative transfer model for retrieval of wheat biophysical parameters from broadband reflectance measurements

This study describes the retrieval of wheat biophysical variables of leaf chlorophyll (Cab), leaf area index (LAI), canopy chlorophyll (CCC), and leaf wetness (Cw) from broadband reflectance data corresponding to IRS LISS-3 (Linear Imaging Self Scanner) sensor by inversion of PROSAIL5B canopy radiative transfer model. Reflectance data of wheat crop, grown under different treatments, were measured by hand-held spectroradiometer and later integrated to LISS-3 reflectance using its band-wise relative spectral response function. Three inversion techniques were used and their performance was compared using different statistical parameters and target diagram. The inversion techniques tried were: a look up table with best solution (LUT-I), a look up table with mean of best 10% solutions (LUT-II) and an artificial neural network (ANN). All the techniques could estimate the biophysical variables by capturing variability in their observed values, though accuracy of estimation varied among the three techniques. Target diagram clearly depicted the superiority of LUT-II over the other two approaches indicating that a mean of best 10% solutions is a better strategy while ANN was worst performer showing highest bias for all the parameters. In all the three inversion techniques, the general order of retrieval accuracy was LAI>Cab>CCC>Cw. The range of Cw was very narrow and none of the techniques could estimate variations in it. In most of the cases, the parameters were underestimated by model inversion. The best identified LUT-II technique was then applied to retrieve wheat LAI from IRS LISS-3 satellite image of 5-Feb-2012 in Sheopur district. The comparison with ground observations showed that the RMSE of LAI retrieval was about 0.56, similar to that observed in ground experimentation. The findings of this study may help in refining the protocol for generating operational crop biophysical products from IRS LISS-3 or similar sensors.