Chemometric technique performances in predicting forest soil chemical and biological properties from UV-Vis-NIR reflectance spectra with small, high dimensional datasets
Chemometric analysis applied to diffuse reflectance spectroscopy is increasingly proposed as an effective and accurate methodology to predict soil physical, chemical and biological properties. Its effectiveness, however, largely varies in relation to the calibration techniques and the specific soil...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Italian Society of Silviculture and Forest Ecology (SISEF)
2016-02-01
|
| Series: | iForest - Biogeosciences and Forestry |
| Subjects: | |
| Online Access: | https://iforest.sisef.org/contents/?id=ifor1495-008 |
| _version_ | 1819227988002078720 |
|---|---|
| author | Bellino A Colombo C Iovieno P Alfani A Palumbo G Baldantoni D |
| author_facet | Bellino A Colombo C Iovieno P Alfani A Palumbo G Baldantoni D |
| author_sort | Bellino A |
| collection | DOAJ |
| description | Chemometric analysis applied to diffuse reflectance spectroscopy is increasingly proposed as an effective and accurate methodology to predict soil physical, chemical and biological properties. Its effectiveness, however, largely varies in relation to the calibration techniques and the specific soil properties. In addition, the calibration of UV-Vis-NIR spectra usually requires large datasets, and the identification of techniques suitable to deal with small sample sizes and high dimensionality problems is a primary challenge. In order to investigate the predictability of many soil chemical and biological properties from a small dataset and to identify the most suitable techniques to deal with this type of problems, we analysed 20 top soil samples of three different forests (Fagus sylvatica, Quercus cerris and Quercus ilex) in southern Apennines (Italy). Diffuse reflectance spectra were recorded in the UV-Vis-NIR range (200-2500 nm) and 22 chemical and biological properties were analysed. Three different calibration techniques were tested, namely the Partial Least Square Regression (PLSR), the combinations wavelet transformation/Elastic net and wavelet transformation/Supervised Principal Component (SPC) regression/ Least Absolute Shrinkage and Selection Operator (LASSO), a kind of preconditioned LASSO. Calibration techniques were applied to both raw spectra and spectra subjected to wavelet shrinkage filtering, in order to evaluate the influence on predictions of spectra denoising. Overall, SPC/LASSO outperformed the other techniques with both raw and denoised spectra. Elastic net produced heterogeneous results, but outperformed SPC/LASSO for total organic carbon, whereas PLSR produced the worst results. Spectra denoising improved the prediction accuracy of many parameters, but worsen the predictions in some cases. Our approach highlighted that: (i) SPC/LASSO (and Elastic net in the case of total organic carbon) is especially suitable to calibrate spectra in the case of small, high dimensional datasets; and (ii) spectra denoising could be an effective technique to improve calibration results. |
| first_indexed | 2024-12-23T10:50:07Z |
| format | Article |
| id | doaj.art-808a2fcbc65f4fadbcac151bd554faa2 |
| institution | Directory Open Access Journal |
| issn | 1971-7458 1971-7458 |
| language | English |
| last_indexed | 2024-12-23T10:50:07Z |
| publishDate | 2016-02-01 |
| publisher | Italian Society of Silviculture and Forest Ecology (SISEF) |
| record_format | Article |
| series | iForest - Biogeosciences and Forestry |
| spelling | doaj.art-808a2fcbc65f4fadbcac151bd554faa22022-12-21T17:49:55ZengItalian Society of Silviculture and Forest Ecology (SISEF)iForest - Biogeosciences and Forestry1971-74581971-74582016-02-019110110810.3832/ifor1495-0081495Chemometric technique performances in predicting forest soil chemical and biological properties from UV-Vis-NIR reflectance spectra with small, high dimensional datasetsBellino A0Colombo C1Iovieno P2Alfani A3Palumbo G4Baldantoni D5Dipartimento di Chimica e Biologia, Università degli Studi di Salerno, v. Giovanni Paolo II 132, I-84084 Fisciano, Salerno (Italy)Dipartimento di Agricoltura Ambiente Alimenti, Università degli Studi del Molise, v. De Sanctis, I-86100 ampobasso (Italy)Consiglio per la Ricerca e la Sperimentazione in Agricoltura (CRA), Centro di ricerca per l’Orticoltura, v. Cavalleggeri 25, I-84098 Pontecagnano, Salerno (Italy)Dipartimento di Chimica e Biologia, Università degli Studi di Salerno, v. Giovanni Paolo II 132, I-84084 Fisciano, Salerno (Italy)Dipartimento di Agricoltura Ambiente Alimenti, Università degli Studi del Molise, v. De Sanctis, I-86100 ampobasso (Italy)Dipartimento di Chimica e Biologia, Università degli Studi di Salerno, v. Giovanni Paolo II 132, I-84084 Fisciano, Salerno (Italy)Chemometric analysis applied to diffuse reflectance spectroscopy is increasingly proposed as an effective and accurate methodology to predict soil physical, chemical and biological properties. Its effectiveness, however, largely varies in relation to the calibration techniques and the specific soil properties. In addition, the calibration of UV-Vis-NIR spectra usually requires large datasets, and the identification of techniques suitable to deal with small sample sizes and high dimensionality problems is a primary challenge. In order to investigate the predictability of many soil chemical and biological properties from a small dataset and to identify the most suitable techniques to deal with this type of problems, we analysed 20 top soil samples of three different forests (Fagus sylvatica, Quercus cerris and Quercus ilex) in southern Apennines (Italy). Diffuse reflectance spectra were recorded in the UV-Vis-NIR range (200-2500 nm) and 22 chemical and biological properties were analysed. Three different calibration techniques were tested, namely the Partial Least Square Regression (PLSR), the combinations wavelet transformation/Elastic net and wavelet transformation/Supervised Principal Component (SPC) regression/ Least Absolute Shrinkage and Selection Operator (LASSO), a kind of preconditioned LASSO. Calibration techniques were applied to both raw spectra and spectra subjected to wavelet shrinkage filtering, in order to evaluate the influence on predictions of spectra denoising. Overall, SPC/LASSO outperformed the other techniques with both raw and denoised spectra. Elastic net produced heterogeneous results, but outperformed SPC/LASSO for total organic carbon, whereas PLSR produced the worst results. Spectra denoising improved the prediction accuracy of many parameters, but worsen the predictions in some cases. Our approach highlighted that: (i) SPC/LASSO (and Elastic net in the case of total organic carbon) is especially suitable to calibrate spectra in the case of small, high dimensional datasets; and (ii) spectra denoising could be an effective technique to improve calibration results.https://iforest.sisef.org/contents/?id=ifor1495-008Elastic NetPLSRSPC/LASSOWaveletsDiffuse Reflectance SpectroscopySample Size |
| spellingShingle | Bellino A Colombo C Iovieno P Alfani A Palumbo G Baldantoni D Chemometric technique performances in predicting forest soil chemical and biological properties from UV-Vis-NIR reflectance spectra with small, high dimensional datasets iForest - Biogeosciences and Forestry Elastic Net PLSR SPC/LASSO Wavelets Diffuse Reflectance Spectroscopy Sample Size |
| title | Chemometric technique performances in predicting forest soil chemical and biological properties from UV-Vis-NIR reflectance spectra with small, high dimensional datasets |
| title_full | Chemometric technique performances in predicting forest soil chemical and biological properties from UV-Vis-NIR reflectance spectra with small, high dimensional datasets |
| title_fullStr | Chemometric technique performances in predicting forest soil chemical and biological properties from UV-Vis-NIR reflectance spectra with small, high dimensional datasets |
| title_full_unstemmed | Chemometric technique performances in predicting forest soil chemical and biological properties from UV-Vis-NIR reflectance spectra with small, high dimensional datasets |
| title_short | Chemometric technique performances in predicting forest soil chemical and biological properties from UV-Vis-NIR reflectance spectra with small, high dimensional datasets |
| title_sort | chemometric technique performances in predicting forest soil chemical and biological properties from uv vis nir reflectance spectra with small high dimensional datasets |
| topic | Elastic Net PLSR SPC/LASSO Wavelets Diffuse Reflectance Spectroscopy Sample Size |
| url | https://iforest.sisef.org/contents/?id=ifor1495-008 |
| work_keys_str_mv | AT bellinoa chemometrictechniqueperformancesinpredictingforestsoilchemicalandbiologicalpropertiesfromuvvisnirreflectancespectrawithsmallhighdimensionaldatasets AT colomboc chemometrictechniqueperformancesinpredictingforestsoilchemicalandbiologicalpropertiesfromuvvisnirreflectancespectrawithsmallhighdimensionaldatasets AT iovienop chemometrictechniqueperformancesinpredictingforestsoilchemicalandbiologicalpropertiesfromuvvisnirreflectancespectrawithsmallhighdimensionaldatasets AT alfania chemometrictechniqueperformancesinpredictingforestsoilchemicalandbiologicalpropertiesfromuvvisnirreflectancespectrawithsmallhighdimensionaldatasets AT palumbog chemometrictechniqueperformancesinpredictingforestsoilchemicalandbiologicalpropertiesfromuvvisnirreflectancespectrawithsmallhighdimensionaldatasets AT baldantonid chemometrictechniqueperformancesinpredictingforestsoilchemicalandbiologicalpropertiesfromuvvisnirreflectancespectrawithsmallhighdimensionaldatasets |