Resting site ecology and microhabitat use of the Mozambique thicket rat (Grammomys cometes) in a riverine Combretum forest

• Main Author: Gebe, Zimkhitha
• Format: Others
• Language: English
• Published: University of Fort Hare 2014
• Online Access: http://hdl.handle.net/10353/d1019768

Very little is known about the resting and spatial ecology of the Mozambique thicket rat, Grammomys cometes, a rodent species distributed in two disjoint populations in South Africa and Mozambique. The present study focused on determining the characteristics and usage patterns of resting sites, as well as the microhabitat use and selection of this species during its activity period. Broadly, I predicted that resting ecology and space use would be affected by environmental, climatic, social and sexual factors. I also hypothesized that predatory risk could affect thicket rat behaviour. Field work took place in the Great Fish River Reserve complex (Eastern Cape Province, South Africa), in a stretch of riverine Combretum forest (length × breadth: ca. 500 × 100 m) located in the western section of the conservation area. Overall, 38 different rats (22 males and 16 females) were trapped and radio-tracked for varying periods between July 2011 and November 2012. Individual rats used a mean of 2.54  1.89 different resting sites (range: 1–9) during each tracking session and resting-site fidelity averaged 85  17%. Overall, 27% of the 131 resting sites identified were artificial wooden nestboxes, 45% were inside branches, 21% were inside trunks, 3% were located on the ground, 2% in bushes, 1% inside dead logs, with one uncertain location (1%). Cape bushwillows Combretum caffrum were the predominant trees (60.32%) used for the resting sites, probably due to their abundance in the forest, and their propension to rot from the inside and provide natural cavities. The mean height of resting sites used by thicket rats was 217  119 cm, and the mean circumference of branches and trunks was 79  35 cm, with no seasonal and sexual variation. In contrast, males exhibited an overall lower percentage usage of individual resting sites than females, probably due to their increased movements during the long mating season. Percentage usage was significantly higher during the cold season, likely for thermoregulatory seasons. Nearly 50% of the resting sites were shared with an average ( SD) of 3.20  1.25 individuals (range: 2–6). However, simultaneous use of resting sites only took place in about 8% of locations. Sharing involved 2–3 males in 90% of the cases. It is suggested that females are solitary but not territorial, while at least some males are more social or tolerant towards each other, and have overlapping ‘home ranges’. These observations could be indicative of a promiscuous mating system. v Microhabitat use and selection were determined by conducting a 5-day trapping session during each season between June 2011 and May 2012. A grid of 96 stations (16 rows  6 lines) was set at the study site, with pairs of traps per station placed at 10-m intervals, at different heights and on different types of vegetative stands. Several variables describing microhabitat structure were recorded seasonally at each station. Thicket rats (38 individuals captured 91 times) were captured mostly at night, with an overall trapping success of 2.16  0.27% over the year. The average height at which rats were trapped (140  65 cm) was significantly higher than the height of traps set in the field (99 ± 75 cm). A large proportion of traps (31.3%) where rats were caught were placed on Combretum caffrum trees, and majoritarily on trunks (46%), tree canopy (19%) and woody lace (19%). Bonferroni Z tests indicated that thicket rats actively avoided trap stations with less than 50% canopy cover as well as areas that had arboreal connections in less than 50% of the directions. Rats also significantly avoided traps that were set in bushes and on the ground, irrespective of whether these were situated in open terrain or surrounded by some vegetation. Generalized Linear Models confirmed that two main predictors (Cover >150 cm and height) had a significant positive effect on the use of traps and on the numbers of visits to, and different animals caught at, trapping stations. All other variables, including connectivity with the surrounding vegetation, tree species, vegetation type and position in which the traps were placed, and tree trunk/branch circumference at trap height, did not have any significant effects. This study confirms that thicket rats are essentially nocturnal, arboreal and prefer dense canopy cover, possibly because this decreases predation risk at the microhabitat level.