Rossel's astronomical work involved far more than observing the sky. He also had to monitor the performance of the timekeepers carried on board, compare their readings with astronomical observations, and assess the errors that could affect calculated positions. The expedition carried several timekeepers made by Ferdinand Berthoud, notably nos. 17, 21, 5 and 28, as well as watches nos. 14 and 10 made by his nephew Louis Berthoud. At the Cape of Good Hope, a watch made by the English clockmaker John Arnold was added to this collection. Rossel had to monitor their rates, compare their readings with astronomical observations, and take their deviations into account when calculating longitude. These instruments represented one of the most advanced techniques then available for determining longitude at sea, but their accuracy had to be continually checked.
Rossel's results formed part of a collective scientific enterprise. Positions determined by astronomical observation provided Beautemps-Beaupré with reference points for constructing his charts and checking the routes followed during hydrographic surveys. Rossel also carried out experiments on terrestrial magnetism. Measurements collected during the voyage helped demonstrate that the intensity of the Earth's magnetic field was not uniform across the globe, but varied notably with latitude.
The calculation notebooks preserved in the French National Archives provide a glimpse into the very workshop of this scientific work: sequences of measurements, corrections, and conversions show how an observation made from a ship could ultimately become a point plotted on a chart. When preparing the expedition for publication, Rossel did more than simply reproduce these observations: he examined the errors that could affect measurements made at sea and proposed methods for reducing their influence on the calculation of latitude and longitude.
The Voyage de Dentrecasteaux, envoyé à la recherche de La Pérouse, published in 1808, was therefore more than the retrospective testimony of a survivor. Rossel established the narrative from d'Entrecasteaux's manuscript journal, which he described as being “written entirely in his own hand”. As the journal ended eleven days before the commander's death, Rossel himself continued the account as far as the frigates' arrival at Surabaya. His work as editor thus preserved and transmitted not only the narrative of the voyage and its scientific results, but also some of the methods by which those results had been obtained.