Optimising Ground Control for River Restoration Monitoring Using an RTK-GNSS UAV and SfM Photogrammetry.

Abstract:

UAVs have revolutionised the availability of high resolution repeat topographic data due to their relatively low-cost and ease of deployment. Consumer-grade Real Time Kinematic Global Navigation Satellite System (RTK-GNSS) equipped UAVs may reduce or eliminate Ground Control Points (GCPs) from SfM photogrammetry surveys, thus removing time-consuming and often hazardous target deployment/collection. However, removal of ground control can substantially reduce the georeferencing accuracy of SfM photogrammetry outputs. Here, we compare GCP configurations using surveys of a 2 x 0.5 km reach of the River Feshie, Scotland with a DJI Phantom 4 RTK UAV.

Checkpoint errors were calculated for seven GCP configuration scenarios. Using no GCPs, 3300 independent GNSS surveyed checkpoints have mean z-axis error -0.010m (RMSE = 0.066 m). Using 5 GCPs gave 0.016 m (RMSE = 0.072 m). Summary error statistics may mask vertical systematic errors, such as doming, in SfM derived Digital Surface Models. Our results from a spatially distributed checkpoint survey do not show such systematic errors, but measuring errors on a spatially distributed set of independent checkpoints is recommended. Restoration projects often use pre-existing imagery over which there is no flight plan control; for new image acquisition campaigns, optimising flight plan design can mitigate commonly experienced systematic errors.

The Feshie has local channel-bar relief of c.1 m and median grain size c.60 mm. Our results imply that an RTK-GNSS UAV can produce acceptable errors with no ground control, alongside spatially distributed independent checkpoints. RTK-GNSS UAVs are versatile for rapid kilometre-scale topographic survey in a range of environments, allowing frequent re-surveys to effectively monitor river and floodplain restoration. We demonstrate this for 1.5 km river corridor of the River Nith, Scotland with surveys of a two-stage channel restoration on completion in mid-2019 and in summer 2020.