Surveying 1,700 hectares of semi-arid terrain in Mexico

To support the design of pluvial runoff and sanitary drainage infrastructure for a large-scale residential development in Mexico, a 1,700-hectare aerial survey was conducted. The team overcame challenges associated with terrain conditions, sparse vegetation and atmospheric factors by combining VTOL UAV technology with ground-based survey methods. The project highlights key considerations from a photogrammetric perspective in terms of data acquisition, processing workflows and the generation of engineering-ready deliverables.

The approximately 1,700-hectare project site is located in a semi-arid region of Querétaro, Mexico. The terrain is characterized by low vegetation density, exposed soil and heterogeneous surface textures, all of which influence both image acquisition and photogrammetric processing. The primary objective was to generate high-resolution topographic data to support hydrological analysis, specifically for pluvial runoff modelling and the design of sanitary drainage systems for a future residential development. These applications require accurate representation of terrain morphology, including natural channels, slopes and water flow paths.

Operational challenges in semi-arid environments

Semi-arid conditions introduce additional constraints such as high reflectance variability, limited natural feature contrast in some areas, and atmospheric interference from dust and haze, all of which can impact data quality and processing reliability. Lighting conditions represented a significant challenge during data acquisition. High solar radiation and minimal cloud cover produced strong contrasts and hard shadows, affecting radiometric consistency between images. Atmospheric dust further reduced image clarity during certain periods, requiring careful selection of flight windows, typically during early morning hours to minimize interference.

Although ground visibility was generally good due to sparse vegetation, localized areas with dense tree cover created occlusions, particularly along natural drainage channels. These obstructions limited the effectiveness of aerial photogrammetry in capturing the true terrain surface in critical zones. Wind conditions also affected platform stability, especially during vertical takeoff and landing (VTOL) transitions. Maintaining consistent flight parameters was essential to ensure uniform ground sampling distance (GSD) and sufficient image overlap.

Accurately identifying terrain variations

Data acquisition was performed using a VTOL uncrewed aerial vehicle (UAV) platform (WingtraOne), selected for its ability to efficiently cover large areas while maintaining high image quality. The survey was executed through a total of seven flights, covering the full extent of the project area. Flights were conducted at an average altitude of approximately 120m above ground level, resulting in a GSD of approximately 4cm per pixel. A total of approximately 4,700 images were acquired. Flight planning ensured adequate forward and side overlap to support robust photogrammetric reconstruction, particularly in areas with low surface texture. At this spatial resolution, small-scale terrain variations relevant for runoff modelling and drainage design could be consistently identified.

To address limitations in areas affected by vegetation cover, a hybrid survey approach was implemented. Ground-based data acquisition using GNSS and total station measurements was carried out in zones where aerial data could not accurately capture terrain features, particularly along drainage channels. This integration ensured a more complete and reliable representation of the terrain, which is critical for hydraulic analysis where small elevation differences can significantly influence design outcomes.

Improved terrain representation

Photogrammetric processing was conducted using Pix4D, following a structured workflow including image alignment, tie point generation and dense point cloud creation. Special attention was given to low-texture areas, where parameter adjustments were required to improve matching performance. Ground control points (GCPs) were incorporated to refine georeferencing accuracy and ensure consistency across the dataset.

The integration of ground-based survey data required additional validation and alignment processes to maintain coherence between datasets. This hybrid approach improved terrain representation in critical areas, particularly for longitudinal profiles and cross-sections used in hydraulic design. Orthomosaic generation included radiometric corrections to compensate for lighting inconsistencies, ensuring usability for engineering applications.

Engineering-ready deliverables and CAD integration

A key component of the project was the transformation of geospatial data into engineering-ready deliverables compatible with CAD-based workflows. The final outputs included digital terrain models, orthomosaics, drainage alignments, cross-sections and detailed hydraulic structures such as culverts. These elements were structured according to industry standards, including layer organization, coordinate systems and annotation conventions. The integration of photogrammetric and ground-based data enabled the generation of reliable inputs for hydrological modelling and infrastructure design, directly supporting decision-making processes in the planning of the residential development.

The importance of hybrid methodologies

As photogrammetric technologies continue to evolve, their integration with traditional surveying methods will remain a key factor in addressing complex real-world projects. This project demonstrates that semi-arid environments present both advantages and limitations for photogrammetric surveys. Therefore, it is important to adapt photogrammetric workflows to the specific challenges of such conditions. While ground visibility is generally favourable, factors such as lighting conditions, atmospheric interference and vegetation occlusions must be carefully managed.

One of the most relevant findings is the importance of hybrid methodologies, because aerial photogrammetry alone may not be sufficient in all scenarios, particularly in areas where vegetation obstructs critical terrain features. The integration of GNSS and total station data provides a robust solution to overcome these limitations. VTOL UAV technology proved highly effective for large-area coverage, offering a balance between efficiency and data quality. The results demonstrate that hybrid approaches significantly improve terrain representation, generating high-quality geospatial data suitable for demanding engineering applications and enhancing the reliability of hydrological analysis and infrastructure design. However, careful planning and environmental assessment remain essential.