Journal Paper Digests 2024 #2
- Physics-Informed Neural Networks for solving transient unconfined groundwater flow
- Towards a cost-effective framework for estimating soil nitrogen pools using pedotransfer functions and machine learning
- An interlaboratory comparison of mid-infrared spectra acquisition: Instruments and procedures matter
- Estimating soil organic carbon content at variable moisture contents using a low-cost spectrometer
Estimating soil organic carbon content at variable moisture contents using a low-cost spectrometer
Research-grade spectrometers such as ASD are widely used in the lab to estimate soil properties, but they are bulky, heavy, and not easily deployable to measure field soils. The newer FT-NIR spectrometers are compact, lightweight, and robust, suitable for developing portable sensors for emerging applications such as field-based soil carbon stock assessment. In this study, we investigated the usefulness of an FT-NIR spectrometer (NanoQuest) for estimating SOC content while correcting for the effect of soil moisture using External Parameter Orthogonalization (EPO), and its performance was compared to that of ASD. To develop EPO transformation, five levels of soil moisture were used at 0, 0.07, 0.13, 0.18, 0.24, and 0.30 g g−1. We tested two modeling approaches: Partial Least Squares Regression (PLSR) and Support Vector Regression (SVR). The results showed that EPO was more effective in correcting for the moisture effect as samples became drier. ASD gave a better performance in estimating SOC with SVR (R2: 0.17 to 0.84, RMSE: 6.1 to 3.9 g C kg−1, bias: −0.3 to 0.1 g C kg−1) after EPO transformation. NanoQuest gave slightly lower, but still satisfactory performance in SOC estimation (R2: 0.17 to 0.70, RMSE: 9.2 to 5 g C kg−1, bias: −0.3 to 0.1 g C kg−1). EPO substantially reduced the bias of the SOC models for both ASD and NanoQuest. This study demonstrates the usefulness of low-cost FT-NIR spectrometers for SOC measurement at varying moisture contents and their great potential for field-deployable soil sensor development.
An interlaboratory comparison of mid-infrared spectra acquisition: Instruments and procedures matter
Diffuse reflectance spectroscopy has been extensively employed to deliver timely and cost-effective predictions of a number of soil properties. However, although several soil spectral laboratories have been established worldwide, the distinct characteristics of instruments and operations still hamper further integration and interoperability across mid-infrared (MIR) soil spectral libraries. In this study, we conducted a large-scale ring trial experiment to understand the lab-to-lab variability of multiple MIR instruments. By developing a systematic evaluation of different mathematical treatments with modeling algorithms, including regular preprocessing and spectral standardization, we quantified and evaluated instruments’ dissimilarity and how this impacts internal and shared model performance. We found that all instruments delivered good predictions when calibrated internally using the same instruments’ characteristics and standard operating procedures by solely relying on regular spectral preprocessing that accounts for light scattering and multiplicative/additive effects, e.g., using standard normal variate (SNV). When performing model transfer from a large public library (the USDA NSSC-KSSL MIR library) to secondary instruments, good performance was also achieved by regular preprocessing (e.g., SNV) if both instruments shared the same manufacturer. However, significant differences between the KSSL MIR library and contrasting ring trial instruments responses were evident and confirmed by a semi-unsupervised spectral clustering. For heavily contrasting setups, spectral standardization was necessary before transferring prediction models. Non-linear model types like Cubist and memory-based learning delivered more precise estimates because they seemed to be less sensitive to spectral variations than global partial least square regression. In summary, the results from this study can assist new laboratories in building spectroscopy capacity utilizing existing MIR spectral libraries and support the recent global efforts to make soil spectroscopy universally accessible with centralized or shared operating procedures.
Towards a cost-effective framework for estimating soil nitrogen pools using pedotransfer functions and machine learning
Globally, the strategic use of nitrogen (N) is important in optimizing economic returns and reducing soil nitrogen losses to the environment. Incorporating reliable estimates of nitrogen (N) mineralized over a growing season (GSN) into N-fertilizer rate prescriptions is critical, but may often lack a direct measurement. For this purpose, Pedotransfer functions (PTFs) of total nitrogen (TN) – representing the stable pool from which N is mineralized and biological nitrogen availability (BNA) – representing the labile pool of N mineralization were used to estimate GSN. GSN was calculated based on TN and BNA results from a soil health database (SHD), which also includes a suite of related soil health parameters (n = 2222). Using a process of recursive feature elimination (RFE) and cost-benefit feature elimination (CBFE), the best predictors of TN, BNA, and GSN were identified using a suite of machine learners (MLs) and regression analysis. For TN, RFE revealed that BNA, active carbon (AC), sand (Sa), and soil organic matter (OM) were the best predictors yielding a Lin’s concordance correlation coefficient (CCC) of 0.80 and a reduction in theoretical cost of 41 % compared to the control. CBFE resulted in AC, soil respiration (SR), clay, Sa, and OM as the most cost-effective predictors of TN with a CCC of 0.79 and a theoretical cost savings 49 % below the cost of using all appropriate soil health parameters in the SHD. With respect to BNA, the best predictors from RFE were aggregate stability (AS), AC, SR, and TN with a CCC of 0.78 and a theoretical cost reduction of 23 %. CBFE retained AC, SR, S, TN, OM and pH as predictors of BNA with a CCC of 0.78 and reduction of 29 % in theoretical cost. Finally, GSN results from RFE identified AS, AC, SR, OM and pH as the best predictors with a 0.82 CCC and 17 % reduction in theoretical cost. CBFE, on the other hand, identified AC, SR, sand, OM, and pH as the most cost-efficient predictors while maintaining a CCC of 0.82 and theoretical cost reduction of 29 %. Of the MLs used for pattern recognition (i.e., cubist, random forest, support vector machine, and stochastic gradient boosting), cubist model outperformed the others for the majority of iterations of the RFE and CBFE processes. The cost-effective framework, and the N-related PTFs developed in this study will greatly enhance our ability to predict of soil N-pool dynamics and the ability to incorporate GSN estimates into N-fertilizer recommendations for producers worldwide. Improvements in predictive strength could be achieved by incorporating climate and soil management practices into PTF development. Another area for improvement and future study would include addition of spatial and landscape variability related to N-measures via digital soil mapping applications.
Physics-Informed Neural Networks for solving transient unconfined groundwater flow
Neural networks excel in various machine learning applications; however, they lack the physical interpretability and constraints crucial for numerous scientific and engineering problems. This limitation hinders their ability to accurately capture and predict complex physical systems’ behavior, potentially yielding inaccurate or unreliable results. Physics-Informed Neural Networks (PINNs) are a class of machine learning models that integrate the power of neural networks with the physical laws governing natural phenomena. PINNs provide an effective tool for solving intricate physical problems, ranging from fluid dynamics to materials science, by incorporating physical constraints into the neural network architecture. PINNs can substantially enhance the accuracy and efficiency of model predictions, even in data-limited situations. This work offers insight into recent developments in the PINN field, including their mathematical formulation and training algorithms, and emphasizes their application in solving transient unconfined groundwater flow. In this context, the phreatic surface acts as a spatiotemporally varying boundary condition, and properly accounting for its position is vital for precise predictions of unconfined groundwater flow and related environmental and engineering applications. The study’s objective is to develop a reliable model for estimating the phreatic surface and the spatiotemporal distribution of piezometric heads in a vertical cross-section of an unconfined aquifer. Two cases are examined: the first involves a homogeneous and isotropic aquifer, while the second comprises a mildly heterogeneous and anisotropic one. The challenges and opportunities arising from this emerging research area are also explored, and essential directions for future research are underscored.