• Regarding the scale of the map, you can find in Roman Large-Scale Mapping in the Early Empire, by O.A.W. Dilke that "The average scale is approximately 1:240 or 1:250, but there is a tendency to make important buildings bigger than they should be, so that the scale variation in the extant parts is from 1:189 to 1:413." The author also notes that the general accuracy of this surveying was unsurpassed in Rome for the next 1500 years.

The scale of 1:240 (on average) is nonetheless noted as a standard for surveying in Rome. In other parts of the Empire, it seems that surveying would have used less detailed scales, as land value would not have justified the same level of accuracy, and the accurate exercise of land deeds was not as crucial as in Rome. In the same paper, there is discussion of the "Orange Cadasters", which are surveyings of agricultural properties around the city of Orange in southern Rhône valley (modern day France), which have a 1:6000 scale. (see . Claude Boisse, Le Tricastin des origines à la chute de l'Empire romain) [by the way, if you like Roman antiquities, this region is famous for its numerous well preserved Romain buildings : Orange has an entire Roman theater which is still used for a famous lyrical music festival in the summer, Arles has an amphitheater still used for corridas, and Nimes has the "Maison carrée" which is an intact Roman temple]

The logic beneath the 1:240 ratio could be devised in the Roman unit system : the centuriation of land plots had a granularity of 20 acti of 120 pedi each, hence 1:240 means that each centuriated plot would be mapped in 10 pedes. More fundamentally, it should be compared with the scales of modern day cadastres : in France, the larger scale available on the public website of the cadastre is 1:200 (it is a relatively practical size indeed).

• Regarding the southeastern orientation, the Forma Urbis was in the "Forum of Peace" which also has an orientation to the southeast for one of its sides). Several major buildings in the city share the same orientation (the Circus Maximus), as well as two of the major roadways (via Appia and via Latina) which leave from the Forum in the southeast direction. It would have made sense that the cartographic depiction of the city would have followed this orientation in coherence with the complex in which it sat, and with the cartographic convention (also found in the Orange Cadasters) to orient the map with an x and y axis corresponding to the local Cardo and Decumanus.

Do note that there are traces of geographic maps on a (much) larger scales from the Roman era, subsisting in the Peutinger Table (a Medieval copy of what is believed to be a copy of a IVth century actualised version of a map made by the order of Augustus' personnal friend and adviser Agrippa) [see Rome's World: The Peutinger Map Reconsidered by Richard J. A. Talbert, Tom Elliott], which is oriented to the north. This is a matter of convenience, as the map represents the entirety of the world known to the romans and the road network of the Empire and its vicinity (including also commercial roads to India, whether by the Red Sea, as described also in the "Périple de la Mer Ethiopique" or through the Parthian Empire, as described in Isidorus of Charax's works such as the Etapes Parthes), and the mediterranean is generally elongated on an east-west axis.