Identifying Plant Part Composition of Forest Logging Residue Using Infrared Spectral Data and Linear Discriminant Analysis.

As new markets, technologies and economies evolve in the low carbon bioeconomy, forest logging residue, a largely untapped renewable resource will play a vital role. The feedstock can however be variable depending on plant species and plant part component. This heterogeneity can influence the physical, chemical and thermochemical properties of the material, and thus the final yield and quality of products. Although it is challenging to control compositional variability of a batch of feedstock, it is feasible to monitor this heterogeneity and make the necessary changes in process parameters. Such a system will be a first step towards optimization, quality assurance and cost-effectiveness of processes in the emerging biofuel/chemical industry. The objective of this study was therefore to qualitatively classify forest logging residue made up of different plant parts using both near infrared spectroscopy (NIRS) and Fourier transform infrared spectroscopy (FTIRS) together with linear discriminant analysis (LDA). Forest logging residue harvested from several Pinus taeda (loblolly pine) plantations in Alabama, USA, were classified into three plant part components: clean wood, wood and bark and slash (i.e., limbs and foliage). Five-fold cross-validated linear discriminant functions had classification accuracies of over 96% for both NIRS and FTIRS based models. An extra factor/principal component (PC) was however needed to achieve this in FTIRS modeling. Analysis of factor loadings of both NIR and FTIR spectra showed that, the statistically different amount of cellulose in the three plant part components of logging residue contributed to their initial separation. This study demonstrated that NIR or FTIR spectroscopy coupled with PCA and LDA has the potential to be used as a high throughput tool in classifying the plant part makeup of a batch of forest logging residue feedstock. Thus, NIR/FTIR could be employed as a tool to rapidly probe/monitor the variability of forest biomass so that the appropriate online adjustments to parameters can be made in time to ensure process optimization and product quality.

Near infrared spectroscopy calibration for wood chemistry: which chemometric technique is best for prediction and interpretation?

This paper addresses the precision in factor loadings during partial least squares (PLS) and principal components regression (PCR) of wood chemistry content from near infrared reflectance (NIR) spectra. The precision of the loadings is considered important because these estimates are often utilized to interpret chemometric models or selection of meaningful wavenumbers. Standard laboratory chemistry methods were employed on a mixed genus/species hardwood sample set. PLS and PCR, before and after 1st derivative pretreatment, was utilized for model building and loadings investigation. As demonstrated by others, PLS was found to provide better predictive diagnostics. However, PCR exhibited a more precise estimate of loading peaks which makes PCR better for interpretation. Application of the 1st derivative appeared to assist in improving both PCR and PLS loading precision, but due to the small sample size, the two chemometric methods could not be compared statistically. This work is important because to date most research works have committed to PLS because it yields better predictive performance. But this research suggests there is a tradeoff between better prediction and model interpretation. Future work is needed to compare PLS and PCR for a suite of spectral pretreatment techniques.

Portable X-ray fluorescence (pXRF) calibration for analysis of nutrient concentrations and trace element contaminants in fertilisers.

With the increasing popularity of local blending of fertilisers, the fertiliser industry faces issues regarding quality control and fertiliser adulteration. Another problem is the contamination of fertilisers with trace elements that have been shown to subsequently accumulate in the soil and be taken up by plants, posing a danger to the environment and human health. Conventional characterisation methods necessary to ensure the quality of fertilisers and to comply with local regulations are costly, time consuming and sometimes not even accessible. Alternatively, using a wide range of unamended and intentionally amended fertilisers this study developed empirical calibrations for a portable handheld X-ray fluorescence (pXRF) spectrometer, determined the reliability for estimating the macro and micro nutrients and evaluated the use of the pXRF for the high-throughput detection of trace element contaminants in fertilisers. The models developed using pXRF for Mg, P, S, K, Ca, Mn, Fe, Zn and Mo had R2 values greater or equal to 0.97. These models also performed well on validation, with R2 values greater or equal to 0.97 (except for Fe, R2val = 0.55) and slope values ranging from 0.81 to 1.44. A second set of models were developed with a focus on trace elements in amended fertilisers. The R2 values of calibration for Co, Ni, As, Se, Cd and Pb were greater than or equal to 0.80. At concentrations up to 1000 mg kg-1, good validation statistics were also obtained; R2 values ranged from 0.97-0.99, except in one instance. The regression coefficients of the validation also had good prediction in the range of 0-100 mg kg-1 (R2 values were from 0.78-0.99), but not as well at lower concentrations up to 20 mg kg-1 (R2 values ranged from 0.10-0.99), especially for Cd. This study has demonstrated that pXRF can measure several major (P, Ca) and micro (Mn, Fe, Cu) nutrients, as well as trace elements and potential contaminants (Cr, Ni, As) in fertilisers with high accuracy and precision. The results obtained in this study is good, especially considering that loose powders were scanned for a maximum of 90 seconds without the use of a vacuum pump.

African soil properties and nutrients mapped at 30 m spatial resolution using two-scale ensemble machine learning.

Soil property and class maps for the continent of Africa were so far only available at very generalised scales, with many countries not mapped at all. Thanks to an increasing quantity and availability of soil samples collected at field point locations by various government and/or NGO funded projects, it is now possible to produce detailed pan-African maps of soil nutrients, including micro-nutrients at fine spatial resolutions. In this paper we describe production of a 30 m resolution Soil Information System of the African continent using, to date, the most comprehensive compilation of soil samples ([Formula: see text]) and Earth Observation data. We produced predictions for soil pH, organic carbon (C) and total nitrogen (N), total carbon, effective Cation Exchange Capacity (eCEC), extractable-phosphorus (P), potassium (K), calcium (Ca), magnesium (Mg), sulfur (S), sodium (Na), iron (Fe), zinc (Zn)-silt, clay and sand, stone content, bulk density and depth to bedrock, at three depths (0, 20 and 50 cm) and using 2-scale 3D Ensemble Machine Learning framework implemented in the mlr (Machine Learning in R) package. As covariate layers we used 250 m resolution (MODIS, PROBA-V and SM2RAIN products), and 30 m resolution (Sentinel-2, Landsat and DTM derivatives) images. Our fivefold spatial Cross-Validation results showed varying accuracy levels ranging from the best performing soil pH (CCC = 0.900) to more poorly predictable extractable phosphorus (CCC = 0.654) and sulphur (CCC = 0.708) and depth to bedrock. Sentinel-2 bands SWIR (B11, B12), NIR (B09, B8A), Landsat SWIR bands, and vertical depth derived from 30 m resolution DTM, were the overall most important 30 m resolution covariates. Climatic data images-SM2RAIN, bioclimatic variables and MODIS Land Surface Temperature-however, remained as the overall most important variables for predicting soil chemical variables at continental scale. This publicly available 30-m Soil Information System of Africa aims at supporting numerous applications, including soil and fertilizer policies and investments, agronomic advice to close yield gaps, environmental programs, or targeting of nutrition interventions.

Comprehensive nutrient analysis in agricultural organic amendments through non-destructive assays using machine learning.

Portable X-ray fluorescence (pXRF) and Diffuse Reflectance Fourier Transformed Mid-Infrared (DRIFT-MIR) spectroscopy are rapid and cost-effective analytical tools for material characterization. Here, we provide an assessment of these methods for the analysis of total Carbon, Nitrogen and total elemental composition of multiple elements in organic amendments. We developed machine learning methods to rapidly quantify the concentrations of macro- and micronutrient elements present in the samples and propose a novel system for the quality assessment of organic amendments. Two types of machine learning methods, forest regression and extreme gradient boosting, were used with data from both pXRF and DRIFT-MIR spectroscopy. Cross-validation trials were run to evaluate generalizability of models produced on each instrument. Both methods demonstrated similar broad capabilities in estimating nutrients using machine learning, with pXRF being suitable for nutrients and contaminants. The results make portable spectrometry in combination with machine learning a scalable solution to provide comprehensive nutrient analysis for organic amendments.

Chemometric modeling of thermogravimetric data for the compositional analysis of forest biomass.

The objective of this study was to investigated the use of chemometric modeling of thermogravimetric (TG) data as an alternative approach to estimate the chemical and proximate (i.e. volatile matter, fixed carbon and ash contents) composition of lignocellulosic biomass. Since these properties affect the conversion pathway, processing costs, yield and / or quality of products, a capability to rapidly determine these for biomass feedstock entering the process stream will be useful in the success and efficiency of bioconversion technologies. The 38-minute long methodology developed in this study enabled the simultaneous prediction of both the chemical and proximate properties of forest-derived biomass from the same TG data. Conventionally, two separate experiments had to be conducted to obtain such information. In addition, the chemometric models constructed with normalized TG data outperformed models developed via the traditional deconvolution of TG data. PLS and PCR models were especially robust in predicting the volatile matter (R2-0.92; RPD- 3.58) and lignin (R2-0.82; RPD- 2.40) contents of the biomass. The application of chemometrics to TG data also made it possible to predict some monomeric sugars in this study. Elucidation of PC loadings obtained from chemometric models also provided some insights into the thermal decomposition behavior of the chemical constituents of lignocellulosic biomass. For instance, similar loadings were noted for volatile matter and cellulose, and for fixed carbon and lignin. The findings indicate that common latent variables are shared between these chemical and thermal reactivity properties. Results from this study buttresses literature that have reported that the less thermally stable polysaccharides are responsible for the yield of volatiles whereas the more recalcitrant lignin with its higher percentage of elementary carbon contributes to the yield of fixed carbon.

Rapid Quantitative Analysis of Forest Biomass Using Fourier Transform Infrared Spectroscopy and Partial Least Squares Regression.

Fourier transform infrared reflectance (FTIR) spectroscopy has been used to predict properties of forest logging residue, a very heterogeneous feedstock material. Properties studied included the chemical composition, thermal reactivity, and energy content. The ability to rapidly determine these properties is vital in the optimization of conversion technologies for the successful commercialization of biobased products. Partial least squares regression of first derivative treated FTIR spectra had good correlations with the conventionally measured properties. For the chemical composition, constructed models generally did a better job of predicting the extractives and lignin content than the carbohydrates. In predicting the thermochemical properties, models for volatile matter and fixed carbon performed very well (i.e., R2 > 0.80, RPD > 2.0). The effect of reducing the wavenumber range to the fingerprint region for PLS modeling and the relationship between the chemical composition and higher heating value of logging residue were also explored. This study is new and different in that it is the first to use FTIR spectroscopy to quantitatively analyze forest logging residue, an abundant resource that can be used as a feedstock in the emerging low carbon economy. Furthermore, it provides a complete and systematic characterization of this heterogeneous raw material.