4.2.1 Global distribution
Over half (59.3%) of all frugivore tracking studies reviewed here were in tropical regions (between 23.436°N and 23.436°S; Fig 2). A further 14.2% were from the sub-tropics and 26.5% were from the southern and northern temperate zones. This reflects latitudinal patterns of higher species richness in tropical compared to temperate regions. However, this pattern was not seen in studies that calculated seed dispersal distances, where only 31% of studies were in tropical regions, and does not reflect difference in animal mediated seed dispersal globally; up to 60 % of temperate plants rely on animal dispersal compared to 90 % of tropical plants (Howe & Smallwood, 1982; Gentry, 1988). This suggests that in the tropics we see a lack of seed dispersal studies using animal movement data, potentially because fieldwork locations and dense forests provide challenges to fieldwork and transmitters (Kays et al, 2011; Monsieurs et al, 2017). However, the uptake of solar powered tags, improvement in battery capacity and remote downloading capabilities may soon rectify this (Byers et al, 2017; Fischer et al, 2018). Our data also suggest that many regions have yet to be explored for the study of frugivore movement, for example Eastern Europe, Central & Northern Africa, Central Asia and much of North America lack tracking studies (Figure 2).
Sites where multiple studies have been conducted tended to be in protected areas and/ or at key field stations. In total, 172 different locations were used for these tracking studies, with almost two thirds (59%) in protected areas, with clustering of multiple studies at long-term research stations (Blanco et al, 2020). This is both a benefit and a limitation; on the one hand, highly studied sites become hotspots for research, with multiple taxa studied in a single location and often with longitudinal datasets (Stouffer, 2020). This provides information on how multiple taxa respond to the same landscape changes and how patterns may differ amongst species. Alternatively, concentrating multiple studies in a few locations means that we make inferences from a handful of intensely studied sites, and lack broader knowledge from diverse locations and landscapes. Most studies we reviewed were in protected areas, but these represent < 10% of terrestrial land surface area, suggesting that we need further studies outside parks and reserves as well as comparative studies in protected areas and neighbouring disturbed habitats to effectively survey a representative sample of habitats