Sighting frequency decreases over two decades in three populations of Testudo hermanni from central Italy

Sighting frequency of tortoises (Testudo hermanni), standardized by the number of person-hours of field effort, was studied between 1992 and 2013 at three different study areas in central Italy. Although the frequency of tortoise sightings fluctuated substantially across years and among study areas, there was a significant declining trend in all the three areas, with GLM analyses showing also a significant interaction between study area and sampling year. The decreasing trend was higher in Castel Fusano than in the other two sites, with yearly frequency of sightings being independent of search effort in all of the study areas. These trends may indicate population declines, because no other explanations seem plausible. Total rainfall from previous autumn to the study period was positively correlated to tortoise sightings in only one locality. The yearly frequency of sightings of juvenile tortoises did not vary significantly among study areas and across years, thus indicating that simple differences in detectability cannot explain the observed patterns. Summer fires and forest overgrowing seem to explain the tortoise decline in two study areas (Castel Fusano and Oriolo) whereas we were not able to identify any specific reason for tortoise decline in the third study area (Manziana).