As lions in North Africa became increasingly persecuted, populations sin the lowlands and coastal forests were eliminated and the population retreated to remoter areas to avoid human contact. Although in parts of Morocco and Algeria lions were not be seen in previously commonly inhabited areas for many decades, the animals were still observed unexpectedly in remoter, less well-explored areas years later.
Lion behaviour in the mountains would be more solitary, somewhat like the Siberian Tiger (A Harland pers.comm) than communal (as per lions in East Africa). However, even under these circumstances lions tend towards social behaviour (Black et al. 2013), so a pair of animals would more likely stay together than disperse. It also seems, from hunting accounts, that in mountainous areas they would use ‘lairs’ usually caves as a base, so that is quite a peculiarity related to the landscape.
Although snow can persist on the highest peaks, the snowfall in the Atlas Mountains tends to be limited to the January February period and the strength of the African sun means that in general although snow cover will occur, the evaporation of a snowfall is quite fast compared to equivalent altitudes in Europe or North America.
Lions were known to range up to 3500m and footprints were used by hunters to track animals in the snow. If an animal was being pursued it would keep to remoter areas and if that meant the snowline it would stay there (see my point about caves below)and would be less likely to drop into lowlands if there was human activity unless there was prey (e.g. livestock). The animals would use mountain passes to track prey and keep to areas clear of human presence. So a lion in the snow would be somewhat transient – due to both the passing nature of snowy conditions in the lower altitudes and the lesser need for the lions themselves to be at higher altitudes other than for transit.
Due to lower density of prey in the North African ecosystem (compared to African savannah or open forest) lions tended to be single, in pairs, breeding pairs or breeding pair with juveniles, rather than the expansive prides of southern Africa (Asiatic lions in India have smaller prides but not as small as North African lions). Single animals would often range on their own, certainly single males (of which at least one photograph exists), but there are also records of single females being shot in high mountain passes. The range of a lion in North Africa would have been quite extensive – we see this today in desert dwelling lions in Namibia for example. In northern forests of Morocco, Tunisia and Algeria in the 19th century some ranges were predictable to the point that locals would know a path that a lion took – and could tell hunters (N Yamaguchi pers comm). However in the Atlas mountains and southern Morocco, certainly in the 20th century, this was much less predictable as the distances animals travelled was far greater.
Examining the extinction of the Barbary lion and its implications for felid conservation. PLoS ONE 8(4):e60174
Yamaguchi, N. and B. Haddane, B. (2002) “The North African Barbary lion and the Atlas lion project,” in International Zoo News, vol. 49, pp. 465–481, 2002