Search

Mining can cause severe disturbances to the soil, which underpins the viability of terrestrial ecosystems. Post-mining rehabilitation relies on measuring soil properties that are critical soil health indicators. Soil visible–near-infrared (vis–NIR) spectroscopy is rapid, relatively accurate and cost-effective for estimating a range of soil properties. Recent advances in infrared detectors and microelectromechanical systems (MEMS) have produced miniaturised, relatively inexpensive spectrometers. Here, we evaluate the spectra from four miniaturised visible and NIR spectrometers, some combinations and a full-range vis–NIR spectrometer to model 29 soil physical, chemical and biological properties used to assess soil health at mine sites. We collected soils from reference undisturbed native vegetation and topsoil stockpiles from seven mines in Western Australia. We evaluated the repeatability of the spectrometers and the accuracy of the spectroscopic models built with seven statistical and machine learning algorithms. The spectra from the visible spectrometer could estimate soil texture (sand, silt, and clay) more accurately than the NIR spectrometers. However, the spectra from the NIR spectrometers produced better estimates of soil chemical and biological properties. By combining the miniaturised visible and NIR spectrometers, we improved the accuracy of their soil property estimates, which were similar to those from the full-range spectrometer. The miniaturised spectrometers and combinations predicted 24 of the 29 soil properties with moderate or greater accuracy (Lin’s concordance correlation, pc ≥ 0.65). The repeatability of the NIR spectrometers was similar to that of the full-range, portable spectrometer. Our results show that the miniaturised NIR spectrometers can produce accurate predictions of soil properties comparable to the (orders of magnitude) more expensive full-range portable system, particularly when combined with a visible range spectrometer. Thus, there is potential to develop rapid, accurate, cost-effective diagnostic capacity to support mine site rehabilitation based on miniaturised spectrometers and deliver significant positive economic and environmental outcomes.

Soil properties, such as organic carbon, pH and clay content, are critical indicators of ecosystem function. Visible–near infrared (vis–NIR) reflectance spectroscopy has been widely used to cost-efficiently estimate such soil properties. Multivariate modelling, such as partial least squares regression (PLSR), and machine learning are the most common methods for modelling soil properties with spectra. Often, such models do not account for the multiresolution information presented in the vis–NIR signal, or the spatial variation in the data. To address these potential shortcomings, we used wavelets to decompose the vis–NIR spectra of 226 soils from agricultural and forested regions in south-western Western Australia and developed a wavelet geographically weighted regression (WGWR) for estimating soil organic carbon content, clay content and pH. To evaluate the WGWR models, we compared them to linear models derived with multiresolution data from a wavelet decomposition (WLR) and PLSR without multiresolution information. Overall, validation of the WGWR models produced more accurate estimates of the soil properties than WLR and PLSR. Around 3.5–49.1% of the improvement in the estimates was due to the multiresolution analysis and 1.0–5.2% due to the integration of spatial information in the modelling. The WGWR improves the modelling of soil properties with spectra.

International audience; Spectroscopic measurements of soil samples are reliable because they are highly repeatable and reproducible. They characterise the samples' mineral-organic composition. Estimates of concentrations of soil constituents are inevitably less precise than estimates obtained conventionally by chemical analysis. But the cost of each spectroscopic estimate is at most one-tenth of the cost of a chemical determination. Spectroscopy is cost-effective when we need many data, despite the costs and errors of calibration. Soil spectroscopists understand the risks of over-fitting models to highly dimensional multivariate spectra and have command of the mathematical and statistical methods to avoid them. Machine learning has fast become an algorithmic alternative to statistical analysis for estimating concentrations of soil constituents from reflectance spectra. As with any modelling, we need judicious implementation of machine learning as it also carries the risk of over-fitting predictions to irrelevant elements of the spectra. To use the methods confidently, we need to validate the outcomes with appropriately sampled, independent data sets. Not all machine learning should be considered 'black boxes'. Their interpretability depends on the algorithm, and some are highly interpretable and explainable. Some are difficult to interpret because of complex transformations or their huge and complicated network of parameters. But there is rapidly advancing research on explainable machine learning, and these methods are finding applications in soil science and spectroscopy. In many parts of the world, soil and environmental scientists recognise the merits of soil spectroscopy. They are building spectral libraries on which they can draw to localise the modelling and derive soil information for new projects within their domains. We hope our article gives readers a more balanced and optimistic perspective of soil spectroscopy and its future. Highlights Spectroscopy is reliable because it is a highly repeatable and reproducible analytical technique. Spectra are calibrated to estimate concentrations of soil properties with known error. Spectroscopy is cost-effective for estimating soil properties. Machine learning is becoming ever more powerful for extracting accurate information from spectra, and methods for interpreting the models exist. Large libraries of soil spectra provide information that can be used locally to aid estimates from new samples.

Near-infrared (NIR) luminescent materials have unparalleled advantages in certain fields, such as anti-counterfeiting. Nd3+ ion doped Gd2O3 near-infrared phosphor was successfully prepared by co-precipitation method and high-temperature solid-phase reaction. The structure and morphology of Gd2O3:Nd3+ cubic phase phosphors were studied in detail. The prepared phosphor was optically analyzed by fluorescence spectrometer. The research results show that with the increase of the Nd3+ doping concentration, the luminous intensity of the near-infrared phosphor in the near-infrared region shows a trend of first increasing and then decreasing. When the Nd3+ ion doping concentration is 1.0 mol%, the obtained phosphors have the best luminous intensity in the near-infrared region under the excitation of 808 nm laser. Gd2O3:Nd3+ phosphor emits near-infrared light due to the intra-4f transitions of Nd3+ ions. The concentration quenching theory is used to prove that the concentration quenching of Gd2O3:Nd3+ belongs to the cross relaxation caused by electric multipole. Due to the unique optical properties of this near-infrared phosphor, it is expected to be further developed in the field of fluorescent materials.

Convolutional neural networks (CNN) for spectroscopic modelling are currently tuned manually, and the effects of their hyperparameters are not analysed. These can result in sub-optimal models. Here, we propose an approach to tune one-dimensional CNN (1D-CNNs) automatically. It consists of a parametric representation of 1D-CNNs and an optimisation of hyperparameters to maximise a model’s performance. We used a large European soil spectroscopic database to demonstrate our approach for estimating soil organic carbon (SOC) contents. To assess the optimisation, we compared it to random search, and to understand the effects of the hyperparameters, we calculated their importance using functional Analysis of Variance. Compared to random search, the optimisation produced better final results and showed faster convergence. The optimal model produced the most accurate estimates of SOC with $$\hbox {RMSE} = 9.67 \pm 0.51$$ RMSE = 9.67 ± 0.51 (s.d.) and $${R}^2 = 0.89 \pm 0.013$$ R 2 = 0.89 ± 0.013 (s.d.). The hyperparameters associated with model training and architecture critically affected the model’s performance, while those related to the spectral preprocessing had little effect. The optimisation searched through a complex hyperparameter space and returned an optimal 1D-CNN. Our approach simplified the development of 1D-CNNs for spectroscopic modelling by automatically selecting hyperparameters and preprocessing methods. Hyperparameter importance analysis shed light on the tuning process and increased the model’s reliability.

Soil fungi play important roles in the functioning of ecosystems, but they are challenging to measure. Using a continental-scale dataset, we developed and evaluated a new method to estimate the relative abundance of the dominant phyla and diversity of fungi in Australian soil. The method relies on the development of spectrotransfer functions with state-of-the-art machine learning and uses publicly available data on soil and environmental proxies for edaphic, climatic, biotic and topographic factors, and visible–near infrared (vis–NIR) wavelengths, to estimate the relative abundances of Ascomycota, Basidiomycota, Glomeromycota, Mortierellomycota and Mucoromycota and community diversity measured with the abundance-based coverage estimator (ACE) index. The algorithms tested were partial least squares regression (PLSR), random forest (RF), Cubist, support vector machines (SVM), Gaussian process regression (GPR), extreme gradient boosting (XGBoost) and one-dimensional convolutional neural networks (1D-CNNs). The spectrotransfer functions were validated with a 10-fold cross-validation (n=577). The 1D-CNNs outperformed the other algorithms and could explain between 45 % and 73 % of fungal relative abundance and diversity. The models were interpretable, and showed that soil nutrients, pH, bulk density, ecosystem water balance (a proxy for aridity) and net primary productivity were important predictors, as were specific vis–NIR wavelengths that correspond to organic functional groups, iron oxide and clay minerals. Estimates of the relative abundance for Mortierellomycota and Mucoromycota produced R2≥0.60, while estimates of the abundance of the Ascomycota and Basidiomycota produced R2 values of 0.5 and 0.58 respectively. The spectrotransfer functions for the Glomeromycota and diversity were the poorest with R2 values of 0.48 and 0.45 respectively. There is no doubt that the method provides estimates that are less accurate than more direct measurements with conventional molecular approaches. However, once the spectrotransfer functions are developed, they can be used with very little cost, and could serve to supplement the more expensive and laborious molecular approaches for a better understanding of soil fungal abundance and diversity under different agronomic and ecological settings.

Leg exoskeletons have gained popularity in clinical settings for gait training in the past decades. However, current devices are restricted in clinical setting due to their complexity and bulkiness. On the contrary, 1-DOF linkage based finger exoskeletons for continuous passive motion (CPM) rehabilitation can be compact and portable. Such concepts have not been exploited in leg exoskeletons. In this paper we attempt to develop a leg exoskeleton based on a 1-DOF linkage for CPM rehabilitation. First dimensional synthesis returns optimal dimensions for a 1-DOF-8-bar-10-joint linkage. Then dynamic simulation estimates the input torque for actuation design. An open-loop control system is also developed. In the end, we achieved a leg exoskeleton that can generate a human-like gait pattern at hip and knee joints. The device can deliver CPM based rehabilitation. Due to its compactness and portability, it may be used outside clinical settings.

Exoskeletons can assist wearers to relearn natural movements when attached to the human body. However, most current devices are bulky and heavy, which limit their application. In this paper, we integrated type and dimensional synthesis to design one degree-of-freedom (DOF) linkages consisting of only revolute joints with multiple output joints for compact exoskeletons. Type synthesis starts from a four-bar linkage where the output link generates the first angular output. Then, an RRR dyad is connected to the four-bar linkage for the second angular output while ensuring that the overall DOF of the new mechanism is 1. A third output joint is added in a similar manner. During each step, dimensional synthesis is formulated as a constrained optimization problem and solved via genetic algorithms. In the first case study, we developed a finger exoskeleton based on a 10-bar-13-joint linkage for a natural curling motion. The second case study presents a leg exoskeleton based on an 8-bar-10-joint linkage to reproduce a natural walking gait at the hip and knee joints. We manufactured the exoskeletons to validate the proposed approach.