Raw Vis-NIR and MIR spectral data and lab measurements from a soil study in Murehwa, Zimbabwe. The dataset supports the research paper 'Multivariate regional deep learning prediction of soil properties from near-infrared, mid-infrared and their combined spectra' published in Geoderma Regional in 2024. The data was authored by Nyawasha, Rumbidzai W and is hosted by CIRAD.
Use Cases
- Predict soil organic carbon content based on spectral reflectance data.
- Predict total nitrogen content based on spectral reflectance data.
- Predict soil texture properties based on spectral reflectance data.
- Train multivariate deep learning models using combined near-infrared and mid-infrared spectra.
- Compare prediction accuracy between different spectral ranges for soil analysis.
Strengths
- Data is directly linked to a published 2024 research paper in Geoderma Regional.
- Includes both raw spectral data and corresponding laboratory measurements.
- Focuses on a specific geographic region (Murehwa, Zimbabwe), providing localized context.
Limitations
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count is unknown, which may limit suitability assessment.
- Description metadata is limited; actual data quality requires manual inspection after download.
Provenance
- Source
- CIRAD Harvested Collection
- Collection Method
- Likely collected via field spectroscopy and laboratory analysis.
- Freshness
- Last updated 2026-05-09 09:10:12; freshness should be verified.
- Geography
- Murehwa, Zimbabwe