Comparison of laboratory versus in situ visible/near‐ (visNIR) and mid‐infrared (MIR) spectroscopy for prediction of various soil properties is required to demonstrate trade‐offs between accuracy and efficiency. Field measurements were made on an arable field in Germany (silt loam Haplic Luvisol) using visNIR (ASD FieldSpec 3 Hi‐Res) and MIR (Agilent Technologies 4300 Handheld FTIR) and material was collected for lab visNIR (Foss XDS Rapid Content Analyzer) and MIR (Bruker‐TENSOR 27) measurements on dried and ground soil and determination of total, labile (>63 μm light), stabilized (>63 μm heavy + <63 μm oxidizable) and resistant organic carbon (OC) content, total nitrogen (N _t ), pH and texture. Partial least squares regression models were calculated for five repeated partitions of the dataset ( n = 238) into training (75%) and test (25%) sets. Lab spectral models outperformed in situ models for total OC (root mean squared error [RMSE] = 0.24–1.0 g kg ⁻¹ ), N _t (RMSE = 0.026–0.088 g kg ⁻¹ ), pH (RMSE = 0.12–0.28) and texture (RMSE = 0.53–1.5%). For both lab and field spectra, the accuracy of visNIR models was comparable or slightly better than MIR for sand, silt and clay. Spectral estimations for labile (RMSE = 0.34–0.47 g kg ⁻¹ ) and stabilized OC (RMSE = 0.41–0.85 g kg ⁻¹ ) were slightly (lab spectra) to substantially (field spectra) inferior to estimations from multiple linear regressions using total OC, N _t , clay and pH as predictors. Variable importance in the projection scores elucidated differences in spectral prediction mechanisms by spectrometer and OC fraction, and found mineral spectral signatures were highly important. For this field‐scale study with 14% median soil gravimetric water content (GWC), the loss of accuracy from lab to field measurement was lower for visNIR than MIR. Analysis of the driest soils (18%), demonstrating the moisture dependence of performance rankings. Highlights: Lab vs field visible/near‐ (visNIRS) and mid‐infrared (MIRS) spectroscopy require comparison for prediction of soil C fractions, N, pH and texture. Lab MIRS prediction of C, N and pH were superior, while texture estimations were comparable or slightly inferior to lab visNIRS. At 14% median soil water content, the loss of accuracy from lab to field measurement was lower for visNIRS than MIRS. The ranking of field visNIRS vs MIRS performance for C and N estimation is moisture dependent.

Infrared spectroscopy in the visible to near‐infrared (vis–NIR) and mid‐infrared (MIR) regions is a well‐established approach for the prediction of soil properties. Different data fusion and training approaches exist, and the optimal procedures are yet undefined and may depend on the heterogeneity present in the set and on the considered scale. The objectives were to test the usefulness of partial least squares regressions (PLSRs) for soil organic carbon (SOC), total carbon (C _t ), total nitrogen (N _t ) and pH using vis–NIR and MIR spectroscopy for an independent validation after standard calibration (use of a general PLSR model) or using memory‐based learning (MBL) with and without spiking for a national spectral database. Data fusion approaches were simple concatenation of spectra, outer product analysis (OPA) and model averaging. In total, 481 soils from an Austrian forest soil archive were measured in the vis–NIR and MIR regions, and regressions were calculated. Fivefold calibration‐validation approaches were carried out with a region‐related split of spectra to implement independent validations with n ranging from 47 to 99 soils in different folds. MIR predictions were generally superior over vis–NIR predictions. For all properties, optimal predictions were obtained with data fusion, with OPA and spectra concatenation outperforming model averaging. The greatest robustness of performance was found for OPA and MBL with spiking with R ² ≥ 0.77 (N), 0.85 (SOC), 0.86 (pH) and 0.88 (C _t ) in the validations of all folds. Overall, the results indicate that the combination of OPA for vis–NIR and MIR spectra with MBL and spiking has a high potential to accurately estimate properties when using large‐scale soil spectral libraries as reference data. However, the reduction of cost‐effectiveness using two spectrometers needs to be weighed against the potential increase in accuracy compared to a single MIR spectroscopy approach.

Previous studies investigating multi-sensor fusion for the collection of soil information have shown variable improvements, and the underlying prediction mechanisms are not sufficiently understood for spectrally-active and -inactive properties. Our objective was to study prediction mechanisms and benefits of model fusion by measuring mid-infrared (MIR) and X-ray fluorescence (XRF) spectra, texture, total and labile organic carbon (OC) and nitrogen (N) content, pH, and cation exchange capacity (CEC) for n = 117 soils from an arable field in Germany. Partial least squares regression models underwent a three-fold training/testing procedure using MIR spectra or elemental concentrations derived from XRF spectra. Additionally, two sequential hybrid and two high-level fusion approaches were tested. For the studied field, MIR was superior for organic properties (ratio of prediction to interquartile distance of validation (RPIQV) for total OC = 7.7 and N = 5.0)), while XRF was superior for inorganic properties (RPIQV for clay = 3.4, silt = 3.0, and sand = 1.8). Even the optimal fusion approach brought little to no accuracy improvement for these properties. The high XRF accuracy for clay and silt is explained by the large number of elements with variable importance in the projection scores >1 (Fe ≈ Ni > Si ≈ Al ≈ Mg > Mn ≈ K ≈ Pb (clay only) ≈ Cr) with strong spearman correlations (±0.57 < r_s < ±0.90) with clay and silt. For spectrally-inactive properties relying on indirect prediction mechanisms, the relative improvements from the optimal fusion approach compared to the best single spectrometer were marginal for pH (3.2% increase in RPIQV versus MIR alone) but more pronounced for labile OC (9.3% versus MIR) and CEC (12% versus XRF). Dominance of a suboptimal spectrometer in a fusion approach worsened performance compared to the best single spectrometer. Granger-Ramanathan averaging, which weights predictions according to accuracy in training, is therefore recommended as a robust approach to capturing the potential benefits of multiple sensors.