Suitability of faecal near-infrared reflectance spectroscopy (NIRS) predictions for estimating gross calorific value
| Autor: | Begoña de la Roza-Delgado, Fernando Vicente, Ana Soldado, Adela Martínez-Fernández, S. Modroño |
|---|---|
| Přispěvatelé: | by the Spanish INIA (Project RTA2011-00135-00-00A), by the Asturias Regional Government, and by European Regional Development Fund (ERDF) |
| Rok vydání: | 2015 |
| Předmět: |
NIR spectroscopy
heating value faeces Accuracy and precision Ecology Near-infrared spectroscopy lcsh:S Cross-validation Livestock Agricultural engineering lcsh:Agriculture Standard error Statistics Calibration Nir spectra Heat of combustion Near infrared reflectance spectroscopy Agronomy and Crop Science Mathematics |
| Zdroj: | Spanish Journal of Agricultural Research; Vol. 13 No. 1 (2015); e0203 Spanish Journal of Agricultural Research; Vol. 13 Núm. 1 (2015); e0203 SJAR. Spanish Journal of Agricultural Research instname Spanish Journal of Agricultural Research, Vol 13, Iss 1, p e0203 (2015) Spanish Journal of Agricultural Research; Vol 13, No 1 (2015); e0203 |
| ISSN: | 2171-9292 1695-971X |
| DOI: | 10.5424/sjar/2015131-6959 |
| Popis: | A total of 220 faecal pig and poultry samples, collected from different experimental trials were employed with the aim to demonstrate the suitability of Near Infrared Reflectance Spectroscopy (NIRS) technology for estimation of gross calorific value on faeces as output products in energy balances studies. NIR spectra from dried and grounded faeces samples were analyzed using a Foss NIRSystem 6500 instrument, scanning over the wavelength range 400-2500 nm. Validation studies for quantitative analytical models were carried out to estimate the relevance of method performance associated to reference values to obtain an appropriate, accuracy and precision. The results for prediction of gross calorific value (GCV) of NIRS calibrations obtained for individual species showed high correlation coefficients comparing chemical analysis and NIRS predictions, ranged from 0.92 to 0.97 for poultry and pig. For external validation, the ratio between the standard error of cross validation (SECV) and the standard error of prediction (SEP) varied between 0.73 and 0.86 for poultry and pig respectively, indicating a sufficiently precision of calibrations. In addition a global model to estimate GCV in both species was developed and externally validated. It showed correlation coefficients of 0.99 for calibration, 0.98 for cross-validation and 0.97 for external validation. Finally, relative uncertainty was calculated for NIRS developed prediction models with the final value when applying individual NIRS species model of 1.3% and 1.5% for NIRS global prediction. This study suggests that NIRS is a suitable and accurate method for the determination of GCV in faeces, decreasing cost, timeless and for convenient handling of unpleasant samples. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|