Comparative Analysis of Field and Laboratory Permeability of Sands at Mancora Beach, Peru

Quiroga K.; Banchon J.; Joselachira; Araujo W. · Civil Engineering and Architecture (2026) · DOI: 10.13189/cea.2026.140520

Resumen

The estimation of the permeability coefficient is a key factor in the design and construction of large-scale engineering works, due to its influence on the stability and performance of foundations. Highly permeable soils can facilitate water infiltration, increasing susceptibility to erosion and deformation processes. In this context, the present study analyzes the permeability of sandy soils at Máncora Beach, located in the Piura region, an area of high touristic relevance. The main objective is to develop a mathematical model to estimate permeability by establishing correlations between in situ permeability and several geotechnical parameters derived from field and laboratory testing. The methodology was based on the collection of field and laboratory data from 20 exploration points distributed across five representative zones. Field tests included in situ permeability tests and density measurements using the driven cylinder method, while laboratory tests comprised grain size analysis, permeability tests, angularity assessment, and specific gravity determination. Based on the results, thirteen independent variables and one dependent variable (in situ permeability) were defined. Simple and multiple correlation analyses were performed to evaluate the relationships among variables. Additionally, a LASSO regression model was implemented using Python to identify the most relevant predictors and mitigate multicollinearity effects. The results indicate that laboratory permeability is the most influential variable in estimating in situ permeability. The final model retained six predictors (laboratory permeability, angularity, fines content, moisture content, specific gravity, and D10) and achieved satisfactory predictive performance (R²= 0.81). A linear regression equation was developed, providing a practical tool for estimating in situ permeability from readily measurable soil properties under conditions similar to those investigated in this study.

Instituciones: Universidad de Piura