Abstract
AbstractThe coefficient of linear extensibility (COLE) is used to classify soils according to their swell–shrink potential, and its estimation is crucial for engineering and agronomic applications. The aims of the study were (a) to develop a visible–near‐infrared spectroscopy (Vis–NIRS, 400–2,500 nm) calibration model to estimate COLE, (b) to compare two model validation approaches (mixed data and country‐wise), and (c) to test if a variable selection method improves the estimation accuracy of the calibration models. For this purpose, partial least square regression (PLSR) was used on the spectra of 53 soil samples from Slovakia and 24 samples from the United States. First, a calibration model based on 70% of the entire dataset (including samples from both locations) was developed and validated with the remaining 30% (mixed data approach). Second, a calibration model based on the Slovakian samples was validated with the U.S. samples (country‐wise approach). Higher predictability for COLE with standardized root mean square error (SMRSE) of 0.099 was obtained for the mixed data approach than for the country‐wise validation with SRMSE of 0.279. Furthermore, using interval PLSR (iPLSR) as a variable selection method did not improve the estimation accuracy of the mixed data approach (SRMSE of 0.099), and rather resulted in a twofold increase in SRMSE (0.560) for the country‐wise validation approach. Overall, the good estimation of COLE from Vis–NIRS was attributed to the high correlation of COLE with clay content and spectrally active clay minerals.
Highlights
When the full spectrum was used for the analyses (PLSR), the coefficient of linear extensibility (COLE) was very well estimated for the calibration dataset of the mixed data approach (R2 = .84, RMSE of cross-validation (RMSECV) = 0.022 cm cm−1, and standardized root mean square error (SRMSE) = 0.108; Figure 1a) and for the country-wise approach (R2 = .88, RMSECV = 0.016 cm cm−1, and SRMSE = 0.076; Figure 2a)
For the interval partial least square regression (PLSR) (iPLSR) variable selection method, the five interval sizes tested showed slight differences in estimation accuracy for the country-wise approach, whereas similar results were obtained for the mixed data approach (Table 2)
We discarded the models with interval sizes of 20, 40, and 100 for the mixed data approach and 40 and 100 for the country-wise approach due to a significantly higher RMSECV and lower R2 compared with the other models
Summary
The investigation of shrinkage and swelling of soils is crucial because it affects the physical conditions of the soil surface that sometimes create large and deep cracks during the dry season. Existing traditional methods to quantify the shrink– swell potential of soils include the rod method, the Georgia volume-change test, and the swell–shrink test (Fityus, Cameron, & Walsh, 2005; Grossman, Brasher, Franzmeier, & Walker, 1968; Simon, Oosterhuis, & Reneau, 1987). There are some practical limitations to these methodologies— for instance, an intact soil core is needed to measure the volume change in wetting and drying conditions, which is expensive to obtain and time consuming for a large number of soils. Pedotransfer functions (PTFs) have been used as an alternative to direct methods to estimate COLE (Tall, Gomboš, Kandra, & Pavelková, 2017), but existing PTFs are often site dependent, and a reliable rapid approach to estimate COLE will be beneficial
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
