Measuring range sizes in a 100-g rodent: mini-GPS are more reliable than transmitters, but the location error reduces reliability

Abstract

Home ranges of free-living mammals have typically been studied via radio-tracking to understand how individuals use their environment. Recently, GPS collars have become popular in large mammals. However, GPS collars are rarely used in small mammals, as they are too heavy, especially when needing coating to protect against gnawing. Here we test the efficiency of mini-GPS collars to measure range estimates compared to the use of radio-collars in a small rodent of 100 g body mass. We equipped 20 bush Karoo rats with mini-GPS loggers and thereafter with radio-transmitters to determine ranges. We validated the accuracy of the mini-GPS loggers by comparing them with the fixes from a handheld GPS and found both to be similar. We estimated range sizes using both traditional methods of Kernel and minimum convex polygon estimates as well as modern methods from movement ecology taking the location error of the mini-GPS into account. Using modern methods led to smaller range estimates, but results were in so far consistent that daily ranges for bush Karoo rats determined using mini-GPS were much larger than home range estimates from radio tracking. Using radio-tracking enabled us to establish the central shelter, while the mini-GPS revealed areas where rats had been observed foraging. We found a distinct location error and therefore suggest using modern approaches from movement ecology which can take this error into account. In sum, mini-GPS revealed more accurate estimates of the ranges than radio-tracking in a small rodent of 100 g body mass.