Joseph Boussinesq (1842–1929): a pioneer of mechanical modelling at the end of the 19th Century

Abstract

Researchers in Mechanics who have never heard the name of Joseph Boussinesq are few. Boussinesq first taught at the Faculty of Sciences in Lille for about fifteen years, then, after being elected to the Academy of Sciences in Paris, taught there for over forty years. From his work, both in Lille and in the Academy of Sciences, he left a great many publications, touching on a variety of subjects, as proof of his many talents. Oddly enough, however, Boussinesq is almost always known to the world of scientific research through only one facet of his many talents. He is little known in France outside some specialized circles where his subjects of research are studied.For those who have some knowledge in the history of mechanical science, it certainly comes as a surprise to see such a phenomenon. Here is a scientist of international renown, who wrote equations, 'Boussinesq's equations', came up with new hypotheses, 'Boussinesq's hypotheses', formulae, 'Boussinesq's formulae' etc. and yet today seems to have vanished from all scientific memory. Boussinesq's name, however, is far from being forgotten. While his area is too specialised to be mentioned in a common dictionary he did have an entry, however, signed by H. Berger and C. Ballot [1] in the Dictionary of Scientific Biography. The town of Saint-André de Sangonis, his birthplace, had a commemoration in his honour on 20 April 1996. The town of Montpellier named a street after him. There is a portrait of Boussinesq at the Lille University of Sciences and Technology (USTL), as seen by a mathematician (M. Parreau [2]). During the last decade three papers about mechanics also commemorated Boussinesq, the scientist: one (W.H. Hager and F. Raemy [3]) deals with his works on turbulence, the two others (R. Zeytounian [4,5]) deal with the Approximation. Finally in 2004, the Netherlands created the Dutch Boussinesq Center for Hydrology under the authority of the Koninklijke Nederlandse Akademie van Wetenschappe (KNAW: Royal Netherlands Academy of Arts and Sciences). Since 2005 the Center has been organizing a series of 'Boussinesq Lectures' given by scientists of world renown dealing with hydrology and soil mechanics.Boussinesq was clearly one of the most underrated scientists that USTL has ever had. It is hoped that a more complete portrait paying homage to his incredibly diverse talents, will at last help restore the rank he deserves.It is often said that mechanics is the most physical field of mathematics, and, conversely, that it is the most mathematical field of physics. This double definition emphasizes one of the identity problems of this science, claimed by both communities, that of mathematics and of physics. Mathematicians see in mechanics a concrete field of application for the often abstract theories they formulate, or they analyse mechanical phenomena to draw concepts out of them in order to reach the same abstraction. An example of this mathematical way of looking at mechanics is the portrait of Boussinesq which M. Parreau made in Lille, and in which he remarks ...

Local energy balance, specific heats and the Oberbeck–Boussinesq approximation

“…The Oberbeck-Boussinesq approximation, so named after the pioneering works by Oberbeck [1] and Boussinesq [2], is the basis of most of the contemporary studies on natural or mixed convection flows. Very interesting historical surveys on the origins of this approximation are available in the recent papers by Zeytounian [3] and Bois [4].…”

Numerical and experimental study of Boussinesq wall horizontal turbulent jet of fresh water in a static homogeneous environment of salt water

“…The applicability of Boussinesq approximation [5,6] is limited to an initial fractional density 0 / a ρρ ∆ of 0.05 [7] where 0 ρ is the density at the source and ρ a is the ambient density, but to generalize this approximation Swain et al [8] suggested a condition of 0 /1 a ρρ ∆ 0 for jets of light fluid injected into an ambient environment of high density. In these cases where Boussinesq approximation is invalid, a density equation must be taken into consideration.…”

Interactive boundary layers in turbulent flow

“…The averaged Navier-Stokes equations are also called Reynolds equations. The numerical solution of these equations requires the use of a turbulence model as suggested by Boussinesq for the first time [2].…”