Environmental Systems Research (Apr 2019)

Experimental validation of a cohesive suspended sediment transport model for two Mexican rivers

  • Juan Antonio García-Aragón,
  • Klever Izquierdo-Ayala,
  • María Mercedes Castillo-Uzcanga,
  • Laura Carrillo-Bibriezca,
  • Humberto Salinas-Tapia

DOI
https://doi.org/10.1186/s40068-019-0139-z
Journal volume & issue
Vol. 8, no. 1
pp. 1 – 10

Abstract

Read online

Abstract Background The south of Mexico is the least developed part of the country but its basins generate more than 60% of the country’s available water. Though disturbances to the rivers caused by development are still low, there is concern about the environmental health of the rivers. The calculation of the sediment load with a focus in fine fraction, where nutrients are found, is a priority. However, models for suspended sediment transport in large rivers are difficult to formulate because they carry a large amount of cohesive sediments, those cohesive sediments form aggregates or flocs with primary particles that are less than 65 μm. The hydrodynamic behavior of flocs depends on their size, density and shape, which differs from that of non-cohesive sediments as theirs depends on their interaction with the water column. A classical model to calculate suspended sediment concentration profiles for steady flow conditions is the Rouse equation, which has been extensively validated for non-cohesive suspended sediment. Some authors have demonstrated that when applied in some large rivers in conjunction with non-cohesive settling velocity models it does not perform very well. The difficulty comes from the fact that most of the suspended sediment charge in large rivers is constituted by cohesive sediments. Results Suspended sediments from Mexico’s two largest rivers Usumacinta and Grijalva, with a mean flow rate near river mouth of 2020 m3/s and 1150 m3/s respectively, were analyzed in a rotating annular flume (RAF). The shear velocity obtained in the field by ADCP was reproduced in the annular flume, the size and shape of flocs were obtained by means of PTV. Settling velocity was also obtained to calibrate a settling velocity model appropriate for cohesive sediments. Conclusions The settling velocity model developed for cohesive sediments in conjunction with the Rouse equation allowed the reproduction of suspended sediment concentration profiles for the rivers Usumacinta and Grijalva. The estimated concentration profiles were compared and validated with the measured concentration profiles in the field. Thus, the model obtained through this research can be used to estimate nutrient delivery to the sea from the largest rivers in Mexico.

Keywords