Google Street View shows promise for virtual street tree surveys

- Google Street View™ was used to conduct a virtual survey of street trees.
- The virtual survey performed well for identifying trees to the genus level.
- Species identification and diameter estimates were less accurate.
- This approach can generate some street tree data more efficiently than field work.

Geospatial technologies are increasingly relevant to urban forestry, but their use may be limited by cost and technical expertise. Technologies like Google Street View™ are appealing because they are free and easy to use. We used Street View to conduct a virtual survey of street trees in three municipalities, and compared our results to existing field data from the same locations. The virtual survey analyst recorded the locations of street trees, identified trees to the species level, and estimated diameter at breast height. Over 93% of the 597 trees documented in the field survey were also observed in the virtual survey. Tree identification in the virtual survey agreed with the field data for 90% of trees at the genus level and 66% of trees at the species level. Identification was less reliable for small trees, rare taxa, and for trees with multiple species in the same genus. In general, tree diameter was underestimated in the virtual survey, but estimates improved as the analyst became more experienced. This study is the first to report on manual interpretation of street tree characteristics using Street View. Our results suggest that virtual surveys in Street View may be suitable for generating some types of street tree data or updating existing data sets more efficiently than field surveys.