Premium
Determination of the threshold velocity of soil wind erosion using a wind tunnel and its prediction for calcareous soils of Iran
Author(s) -
Rezaei Mahrooz,
Mina Monireh,
Ostovari Yaser,
Riksen Michel J. P. M.
Publication year - 2022
Publication title -
land degradation and development
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.403
H-Index - 81
eISSN - 1099-145X
pISSN - 1085-3278
DOI - 10.1002/ldr.4309
Subject(s) - pedotransfer function , environmental science , soil science , soil water , aeolian processes , calcareous , mean squared error , coefficient of determination , wind speed , correlation coefficient , soil test , linear regression , geotechnical engineering , hydrology (agriculture) , mathematics , statistics , meteorology , geology , hydraulic conductivity , geography , paleontology , geomorphology
Determination of the threshold velocity (TV) is a crucial step for wind erosion evaluation. Due to the difficulties of direct field measurements, pedotransfer functions (PTFs) and easily measurable soil properties could be used to save time and cost in predicting TV. Therefore, the present study was conducted to predict the TV using PTFs and to assess its influential parameters for calcareous soils of Fars Province, southern Iran. To this end, the TV was measured by a portable wind tunnel at 72 locations in different land uses and soil types across the study site. Various physicochemical and mechanical soil properties were used to develop six PTFs using multiple linear regression. Results showed that the TV varied from 3.0 m s −1 in poor rangelands to 12.83 m s −1 in saline lands. Soil surface shear strength (SS) with a correlation coefficient of 0.85 was the most influential parameter affecting the TV, followed by aggregate mean weight diameter (MWD). Results of the predictive models revealed that PTF 5, which was developed using SS and penetration resistance (PR; R 2 = 0.86, RMSE = 0.85 m s −1 ), and PTF 6, which was developed using MWD and PR ( R 2 = 0.81, RMSE = 1.07 m s −1 ), had the highest performance for predicting the TV. PTF 5 was selected as the final model for predicting the TV since it only needed easily measurable soil properties without soil sample collection. We concluded that the use of PTFs could be an applicable alternative way to predict the TV, particularly at large scales.