Surveying complex tidal flats in the Yangtze Estuary presents significant challenges due to dynamic sediment movements, dense vegetation coverage, and vast intertidal zones. Traditional surveying methods like GNSS-RTK measurements struggle with efficiency and coverage in these environments. Our research explores the capabilities of surveying drone LiDAR systems to overcome these limitations. Surveying UAV platforms equipped with LiDAR sensors offer rapid, high-resolution topographic mapping capabilities that are revolutionizing coastal geomorphological studies.
Surveying drone LiDAR systems integrate three core components: LiDAR sensors, GNSS positioning, and IMU navigation. The fundamental measurement principle involves calculating target coordinates through laser pulse returns. When a laser pulse from the surveying UAV reaches a target point P, the coordinates are derived through spherical coordinate transformation:
$$x_s = S \cos\theta \cos\alpha$$
$$y_s = S \cos\theta \sin\alpha$$
$$z_s = S \sin\theta$$
where $S$ represents the distance between the surveying UAV and target, $\theta$ denotes the vertical angle between the laser beam and nadir direction, and $\alpha$ indicates the horizontal scanning angle. The surveying UAV’s integrated GNSS-IMU system provides precise platform positioning and orientation data at frequencies exceeding 200Hz, enabling centimeter-level georeferencing accuracy for point cloud data.
Our methodological approach involved strategic flight planning across diverse Yangtze Estuary environments. Surveying UAV missions followed these operational parameters:
| Parameter | Configuration | Purpose |
|---|---|---|
| Flight Altitude | 200m AGL | Optimize point density & coverage |
| Flight Speed | <10m/s | Ensure point cloud density >80pts/m² |
| Overlap | 60% along-track 30% cross-track |
Ensure complete coverage & data redundancy |
| Sensor | RIEGL VUX-240 | 1064nm wavelength, 1.5MHz pulse rate |
| Ancillary Data | 24-bit RGB imagery | Feature classification & verification |
Data processing involved sophisticated point cloud filtering algorithms to address vegetation penetration challenges. The cloth simulation filtering (CSF) technique proved most effective for distinguishing ground returns in vegetated tidal flats. We calculated point cloud penetration rates using:
$$P_p = \frac{N_g}{N_t} \times 100\%$$
where $P_p$ represents penetration percentage, $N_g$ denotes ground-classified points, and $N_t$ indicates total points in the segment. For heavily vegetated areas, we implemented multi-flight campaigns to increase effective ground point density.
Validation involved 54 GNSS-RTK checkpoints distributed across three test zones. The vertical accuracy assessment revealed significant seasonal variations in surveying UAV performance:
| Survey Area | Season | Checkpoints | Mean Error (m) | RMSE (m) | Max Error (m) |
|---|---|---|---|---|---|
| North Channel Dikes | Summer | 18 | 0.024 | 0.079 | 0.128 |
| North Channel Dikes | Winter | 18 | -0.017 | 0.093 | 0.138 |
| Jiuduansha Shoal | Winter | 20 | -0.021 | 0.103 | 0.137 |
Analysis confirmed vegetation as the primary error source for surveying drone measurements. Canopy height difference (CHD) significantly impacted accuracy according to the relationship:
$$\Delta z = 0.87 \times \text{CHD} + 0.12$$
where $\Delta z$ represents the elevation error and CHD is the canopy height differential. Winter operations showed 32% improvement in point cloud penetration rates due to seasonal vegetation dieback. The surveying UAV system demonstrated distinct advantages over conventional techniques:
| Metric | Surveying UAV LiDAR | GNSS-RTK | Photogrammetry |
|---|---|---|---|
| Coverage Rate | 15 km²/day | 0.8 km²/day | 12 km²/day |
| Vegetation Penetration | Partial | Full | None |
| Vertical Accuracy | 0.08-0.15m | 0.02-0.05m | 0.15-0.30m |
| Feature Classification | Good | Excellent | Poor |
Through multiple surveying UAV campaigns, we developed optimized protocols for Yangtze Estuary applications. Critical recommendations include conducting surveys during winter months when vegetation density decreases by approximately 40%, implementing multi-angle LiDAR scanning for dense Spartina alterniflora areas, and integrating GNSS-RTK validation transects within UAV survey blocks. These surveying drone techniques achieved CH/T 8023-2011 accuracy standards for 1:500 scale mapping despite challenging intertidal conditions.
Our conclusions establish clear operational guidelines for tidal flat monitoring. Surveying UAV LiDAR delivers optimal results when deployed during November-March periods, with flight lines spaced at 60% overlap across vegetation boundaries. For perennial vegetation zones, dual-flight missions increase ground point density by 70%. The integration of surveying drone data with GNSS-RTK validation and RGB imagery creates comprehensive digital elevation models that accurately represent the complex Yangtze Estuary morphodynamics. This approach provides a transformative solution for large-scale coastal monitoring programs where traditional survey methods prove logistically challenging.