Habitat use and activity patterns of three Eulemur species in the forest and restoration areas of Ranomafana and Ankafobe, Madagascar

Abstract: Anthropogenic deforestation is a major threat to forest environments, as it leads to habitat loss and fragmentation. Many species suffer from these consequences due to reduced resource availability or connectivity. Forests in Madagascar have drastically been reduced in the past few centuries. There are many projects that focus on forest rehabilitation and restoration but these projects often face challenges. Species-habitat relationships can be specific or spatially variable, and habitat traits that local species require are often not considered in restoration planning. Knowledge on how the species cope with a changing environment is lacking. In Madagascar, frugivorous lemurs can have a large impact on forest regeneration due to their role as seed dispersers, and habitat requirements of these species should be considered in restoration planning. Species of the genus Eulemur have shown flexible behavior and potential to cope with environmental changes. In this study, I monitored three Eulemur species at two sites in Madagascar; Ranomafana (E. rubriventer and E. rufifrons) and Ankafobe (E. fulvus), which both included forest and restoration areas, as well as a forest fragment at Ankafobe that has recently been burnt. Using arboreal and terrestrial motion sensor camera traps and occupancy modelling, I tested if these species occurred in the disturbed areas. I also compared their locomotion and activity patterns between the habitat types. I tested if they increased terrestrial locomotion as a possible response to reduced connectivity in the disturbed restoration areas, and if they showed a shift in their activity pattern. Due to the less dense vegetation structure in the disturbed areas the lemurs might be more exposed and therefore show a larger proportion of activity at night, possibly to avoid diurnal predators. Although I detected none of the species at the restoration areas, the results suggested that E. rubriventer and E. rufifrons do occur in the restoration areas, but were not detected due to their small detection probability. These two species used almost exclusively arboreal behavior, with one exception where I detected E. rufifrons on the ground. E. fulvus does not seem to use the restoration areas at Ankafobe. At Ankafobe the restoration areas were very young and contained a very small number of tree species. These areas therefore seemed to not offer sufficient resource availability for E. fulvus to use these areas. E. fulvus occurred in the burnt area. There, the species frequently used terrestrial locomotion, but also in the forest, which could be linked to the absence of terrestrial predators or the vegetation structure, as the availability of horizontal branches was limited, and terrestrial locomotion might have been more efficient. All three species showed cathemeral activity in the forest with a peak of activity in the morning and the afternoon, as well as a few detections during the night. E. fulvus showed a similar pattern in the burnt area but with a more distributed activity during the day. This could be a response to the decrease in food availability after the fire. Overall, the results suggested that the three species show potential to use disturbed areas and are able to adjust their behavior to changes in the environment, which would be beneficial for forest regeneration due to their role as seed dispersers.