Document Type : Research Paper
Soil texture, defined by the percentage of sand, silt, and clay, is one of the most fundamental and frequently measured properties in soil science laboratories. The hydrometer method, based on Stokes' law, is widely used for texture analysis due to its simplicity and low cost. However, in Iran, various hydrometer reading protocols—including two-reading, four-reading, and complete reading methods—are used interchangeably without prior validation against a reference method. Two-reading methods (e.g., 40 sec & 2 hours) are common because they are fast and simple, but their accuracy has been questioned. The complete reading method (multiple readings over 48 hours) is accurate but time-consuming and expensive.
This study aimed to: (1) compare common hydrometer reading methods with the complete reading method as a reference, (2) evaluate the accuracy of different methods for determining sand, silt, and clay percentages in predominant Iranian soils, (3) identify the most suitable method considering time, cost, and accuracy, and (4) determine the actual settling time for the sand-silt and silt-clay boundaries based on the USDA system.
Eighty-nine soil samples were collected from 13 provinces of Iran (Zanjan, Qazvin, Alborz, Tehran, Qom, Arak, Isfahan, Khorramabad, Fars, Khuzestan, Kerman, Mazandaran, and Golestan). Samples were selected to cover a wide range of soil textures (sandy, loamy, silty, and clayey) representative of predominant Iranian soils. After air-drying and passing through a 2-mm sieve, 40 g of soil was dispersed with sodium hexametaphosphate. The hydrometer (ASTM 152H) was used to measure suspension density at multiple times: 30, 40, 60, 120, 300, 600, 900, 1800, 3600, 5400, 7200, 21600, 23400, 25200, 28800, 86400, and 172800 seconds. Using these readings, six different methods were compared: (a) complete reading method (reference, using semi-logarithmic interpolation), (b) four-reading method (30 & 60 sec for sand; 1.5 & 24 h for clay), (c) two-reading method (40 sec & 8 h), (d) two-reading method (40 sec & 7 h), (e) two-reading method (40 sec & 6.5 h), and (f) two-reading method (40 sec & 2 h – Bouyoucos method). Sand content was also verified by wet sieving through a 53-µm sieve. Statistical analysis included ANOVA and Duncan's test (SPSS26), and accuracy was assessed using MAE, MBE, RMSE, and NRMSE.
The accuracy of the 40-second and 8-hour method was the highest among simplified methods, correctly identifying soil texture class in 90% of samples, followed by the four-reading method (89% accuracy). In contrast, the conventional 40-second and 2-hour method (Bouyoucos) had the lowest accuracy (only 36% agreement with the complete reading method). Clay content was overestimated by all simplified methods compared to the complete reading method, while silt content was underestimated except by the 40-sec & 8-h method. The mean absolute error (MAE) for clay was 1.99% for the 40-sec & 8-h method, 2.06% for the four-reading method, but 12.11% for the 40-sec & 2-h method. The NRMSE for clay was 8.35%, 8.94%, and 36.66%, respectively. ANOVA and Duncan's test showed no significant difference (P<0.01) between the complete reading, four-reading, 40-sec & 8-h, and 40-sec & 7-h methods for clay, silt, and sand values. However, the 40-sec & 6.5-h and 40-sec & 2-h methods showed significant differences (P<0.01) from the reference. The actual settling time for the sand-silt boundary (0.05 mm) averaged 49 seconds (range 30–69 sec) and for the silt-clay boundary (0.002 mm) averaged 11 hours (range 8.5–12.5 h), depending on soil texture. Finer-textured soils required shorter reading times, approaching 40 sec and 8 h.
The simplified 40-second and 8-hour hydrometer method, followed by the four-reading method, provides high accuracy (90% and 89% textural class agreement, respectively) and is statistically comparable to the complete reading method. These methods save considerable time (reducing from 48 hours to 8 or 24 hours) and laboratory costs without significant loss of accuracy. The conventional 40-second and 2-hour method (Bouyoucos) is not recommended for studies requiring precise texture data, as it causes substantial overestimation of clay and shifts texture classes toward heavier textures. Depending on the study objectives and sample number, the 40-second & 8-hour or four-reading methods are reliable alternatives to the complete reading method for determining soil texture in Iranian soils.
The study was funded by the Soil and Water Research Institute, Karaj, Iran.
Roya Toluee: Conceptualization, methodology, software, formal analysis, investigation, resources, data curation, writing—original draft preparation, writing—review and editing, original idea of study, supervision.
Karim Shahbazi: Conceptualization, methodology, investigation, writing—review and editing, original idea of study, supervision, project administration.
Mohammad Hossein Mohammadi: Conceptualization, methodology, investigation, writing—review and editing, supervision.
Meisam Rezaei: Conceptualization, investigation, writing—review and editing.
There is no use of any type of artificial intelligence-based technology in this paper.
Data available on request from the authors.
The authors gratefully acknowledge the financial support provided by the Soil and Water Research Institute, Karaj, Iran. This support was instrumental in facilitating the successful completion of the present research.
The authors avoided data fabrication, falsification, plagiarism, and misconduct.
The authors declare no competing interests.
