The drip irrigation paradox in an arid irrigated system: a soil quality index reveals secondary salinization and nutrient accumulation in Southern Peru

Abstract

In arid regions, the conversion of deserts into agricultural land through irrigation is critical for food security but may intensify soil salinization processes. In this context, integrated and spatial soil quality assessment is essential for guiding sustainable management strategies. In this study, we developed a spatially explicit weighted Soil Quality Index (SQIw) and evaluated the effects of irrigation system (drip vs. surface) on soil salinity and fertility within an emerging irrigation scheme in the Peruvian coastal desert. A total of 930 composite soil samples (0–20 cm depth), each composed of 10 subsamples collected within 1-ha parcels, were analyzed for texture, pH, electrical conductivity (EC1:5), organic matter (OM), available phosphorus (Pav), available potassium (Kav), exchangeable sodium (exNa), and carbonates. Principal Component Analysis identified a Minimal Data Set (sand, EC1:5, Kav, OM, pH) explaining 76.8% of the total variance. Indicators were transformed into dimensionless scores (0.1–1.0) using threshold-based scoring functions and weighted according to PCA-derived communalities. Spatial patterns were characterized using semivariogram modeling and Ordinary Kriging, with 10-fold cross-validation confirming unbiased spatial predictions (ME ≈ 0; RMSE = 0.100), though the coefficient of determination remained limited (R² = 0.246), consistent with the moderate nugget effects observed across soil properties. The SQIw revealed that 46.7% of the interpolated area fell into the "Poor" category, primarily associated with high salinity areas (EC1:5 > 8 dS m⁻¹) and the predominance of coarse-textured soils, whereas the "Acceptable" category accounted for 53.1% of the evaluated area. Additionally, soils under drip irrigation were associated with significantly higher EC, Kav, and Pav values than surface irrigation, consistent with inadequate leaching fractions, though differences in land development age and management history among sampled fields should be considered when interpreting these results. The SQIw effectively discriminated spatial gradients of soil quality, demonstrating that management-induced improvements remain constrained by inherent edaphic limitations. In naturally saline soils, drip irrigation without specific management practices (e.g., salt leaching) poses a risk of secondary salinization. The proposed methodological framework provides a data-driven tool to support differentiated soil management in arid regions worldwide.