Data for: Soil Quality Assessment in Tropical Acid Sulfate Tidal Swamplands Using PCA-Based Minimum Data Set and UAV Multispectral Imagery for Site-Specific Rice Management"
Description
This dataset contains soil chemical property data collected from eight composite sampling points (n = 8) across four representative tidal swampland sites in Barito Kuala Regency, South Kalimantan, Indonesia (3°11'–3°25' S; 114°35'–114°51' E). Data were collected in December 2024 under contrasting wet land and dry land conditions at four villages: Cahaya Baru, Jejangkit Muara, Jejangkit Sempurna, and Jejangkit Timur. Soil samples were collected from the surface horizon (0–20 cm depth) using a Belgian auger. The following soil chemical parameters were measured in the laboratory: Soil Organic Carbon (SOC, %), Total Nitrogen (N-total, %), Total Phosphorus (P₂O₅, %), Total Potassium (K₂O, %), Total Iron (Fe-total, %), Calcium (Ca, %), Magnesium (Mg, %), Pyrite content (FeS₂, %), Soluble Iron (Fe-sol, ppm), Available Phosphorus (P₂O₅-avail, ppm), Soluble Sulfate (SO₄-sol, ppm), Soil pH in water (pH H₂O, 1:5), and Soil pH in KCl solution (pH KCl, 1N 1:5). The C/N ratio was derived from SOC and N-total values. This dataset comprises four components: (1) Laboratory analysis results of 13 soil chemical indicators measured from eight sampling points under wet and dry land conditions, including raw values used to construct the Soil Quality Index (SQI); (2) Principal Component Analysis (PCA) computation data, including factor loadings, eigenvalues, percentage of variance explained, Minimum Data Set (MDS) selection criteria, indicator weights (Wi), scoring results, and final SQI values per sampling point; (3) GPS coordinate data for all eight soil sampling points, including site name, land condition, soil classification, overflow typology, and geographic coordinates (latitude, longitude, and elevation); and (4) UAV-derived multispectral imagery in GeoTIFF format (.tif), consisting of orthorectified reflectance mosaics for four spectral bands (Green 560 nm, Red 650 nm, Red Edge 730 nm, Near Infrared 860 nm) acquired using the DJI Mavic 3 Multispectral (DJI Mavic 3M) UAV at each of the four study sites, along with extracted spectral index values (NDVI, NDRE, GNDVI, LCI, and OSAVI) at each sampling point. This dataset is associated with the manuscript: "Soil Quality Assessment in Tropical Acid Sulfate Tidal Swamplands Using PCA-Based Minimum Data Set and UAV Multispectral Imagery for Site-Specific Rice Management," submitted to Catena (Elsevier).
Files
Steps to reproduce
1. Soil samples were collected from the surface horizon (0–20 cm) using a Belgian auger at eight composite sampling points across four tidal swampland sites in Barito Kuala Regency, South Kalimantan, Indonesia, under wet and dry land conditions (December 2025). 2. Thirteen soil chemical parameters were analyzed following standard laboratory protocols: SOC by Walkley-Black wet oxidation; N-total by semi-micro Kjeldahl; P₂O₅, K₂O, Fe-total, Ca, and Mg by wet digestion H₂SO₄ and AAS/flame photometry; pyrite (FeS₂) by 50% H₂O₂ oxidation gravimetry; Fe-sol by NH₄OAc extraction and AAS; P-avail by Bray I method; SO₄ sol by BaSO₄ turbidimetry; pH H₂O and pH KCl by pH meter (1:5 suspension). 3. PCA was performed on 14 soil indicators (including C/N ratio) using SPSS v.26.0. Four principal components (eigenvalue ≥ 1) explaining 90.69% of total variance were retained. MDS indicators (Mg, K₂O, Fe-total, FeS₂, pH KCl) were selected based on factor loadings and correlation criteria. SQI was calculated using variance-based weights and linear scoring functions. 4. UAV multispectral imagery was acquired using DJI Mavic 3 Multispectral (Green 560 nm, Red 650 nm, Red Edge 730 nm, NIR 860 nm) at 120 m altitude. Images were processed in Agisoft Metashape using SfM workflow. Spectral indices (NDVI, NDRE, GNDVI, LCI, OSAVI) were extracted from a 3×3 pixel window at each GPS-referenced sampling point.