研究目的
To assess the use of diffuse reflectance spectroscopy (DRS) to generate an efficient prediction model for mangrove soil organic carbon (SOC) quantification and to stimulate future studies to expand the use of this technique for analysis of mangrove soils.
研究成果
The results provide evidence that reflectance spectroscopy can be efficiently used for determining organic carbon in mangrove soils, demonstrating greater accuracy and speed compared with traditional analyses. The full-range spectrum (350–2500 nm) generated a better model of prediction when compared with models using a reduced number of wavelengths. However, further studies are necessary before implementing DRS considering the variability among the mangroves worldwide.
研究不足
The study acknowledges the high variability in SOC content and other soil components in mangrove soils worldwide, suggesting the need for further studies to examine the use of DRS for mangrove soils with higher SOC content to generate efficient models for better quantifications in the future.
1：Experimental Design and Method Selection:
The study utilized diffuse reflectance spectroscopy (DRS) in the visible, near-infrared, and shortwave infrared (Vis-NIR-SWIR) regions to predict SOC in mangrove soils. Partial least square regression (PLSR) was used for deriving regression models from spectral information and reference analytical data.
2：Sample Selection and Data Sources:
Soil samples were collected from three mangrove forests in northeastern Brazil, characterized by distinct environmental settings and anthropogenic impacts. A total of 72 soil samples were collected from different depths.
3：List of Experimental Equipment and Materials:
A FieldSpec 4 Hi-Res spectroradiometer (Analytical Spectral Devices) was used for spectrum acquisition. Samples were prepared by drying and sieving before spectral readings.
4：Experimental Procedures and Operational Workflow:
Reflectance spectra data were acquired in a dark room using a Spectralon panel as reference. Three soil spectra were obtained per sample, with the final spectrum being the mean reflectance for each wavelength.
5：Data Analysis Methods:
PLSR analysis was performed using The Unscrambler (Camo Software AS). Models were evaluated based on R2, RMSE values, and residual predictive deviation (RPD).
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
