Optimal Parameter Configuration of UAV for Measuring Gully Morphology in Black Soil Region of Northeast China

[Objective] Severe gully erosion in the black soil region of Northeast China is continuously eroding arable land resources, posing a threat to the country's food security. The development of unmanned aerial vehicle (UAV) technology provides an effective way for monitoring gully erosion, but the higher the resolution of data obtained, the longer it takes. The challenge lies in configuring UAV parameters to simultaneously meet accuracy requirements and enhance aerial photography efficiency. [Methods] The Hebei catchment in the rolling hills region of Northeast China was chosen as the study area. Using actual cross-section measurement data as a validation value, the accuracy of gully parameters extracted from UAV data at different resolutions, flight directions, and types was assessed to explore the suitable conditions for various UAV parameter configurations. [Results] (1) Compared with field measurement data, data at a 1 cm resolution extracted the gully parameters with the highest accuracy, the average percentage error for all parameters was less than 5.0%, suitable for monitoring typical gully development processes. Data extracted at 3 cm and 5 cm resolutions had an average percentage error of less than 10.0% for gully width, and the average error increased as gully depth decreased. For gullies deeper than 1 m, the average percentage error was less than 10.0%, suitable for rapid regional sampling surveys. Data extracted at resolutions of 8 cm and 10 cm had average percentage errors greater than 40.0% for gully depth and cross-sectional area, suitable for extracting gully distribution locations and planar parameters. (2) Although fixed-wing UAV obtaining data slightly outperformed in extracting two-dimensional gully features, multirotor UAV using oblique photogrammetry were better in extracting three-dimensional gully features. (3) The average percentage errors for gully depth and cross-sectional area extracted from single-direction oblique photogrammetry data were 1.7 and 1.9 times those from cross-direction data. UAV data obtained from cross-direction flights provided higher accuracy in gully parameter extraction and richer details. [Conclusion] Setting UAV parameters to a resolution of less than 5 cm and obtaining cross-direction oblique aerial photography data can meet the accuracy requirements for monitoring gully morphology in the black soil region of Northeast China.