Assessment of three methods to determine particle size distribution analysis in soils of Costa Rica

Abstract

Introduction. Accurate determination of soil particle size distribution (PSD) is essential in soil taxonomy, soil physics, and soil management. Although the pipette method is precise, it is impractical for routine laboratory analyses involving large sample volumes. Objective. To evaluate and propose an efficient and practical method for soil texture determination in soils from Costa Rica. Materials and methods. A total of 24 samples from 12 representative soil profiles were analyzed using two soil sample dispersion procedures: 2) industrial blender, and 2) horizontal shaker; and three hydrometer-based methodologies: 1) Bouyoucos, 2) Bouyoucos corrected by temperature (Bouyoucos T), and 3) Bouyoucos modified (Bouyoucos M), compared against the pipette reference method, for a total of 144 analyses. Performance was assessed through linear regression, RMSE, Willmott’s concordance index, and Nash–Sutcliffe efficiency. Results. The horizontal shaker achieved superior particle dispersion compared to the industrial blender. Among the tested methodologies, the modified Bouyoucos method (Bouyoucos M) exhibited the highest agreement with pipette-derived values (R² > 0.9 for clay fraction), particularly when correcting for sample gravimetric moisture. Direct sand determination by sieving did not improve accuracy over the modified method. Conclusions. The modified Bouyoucos method, combined with horizontal shaker dispersion and moisture correction, provides an effective and reliable alternative to the reference pipette method and is recommended for routine soil texture analysis in Costa Rican laboratories.