Creating digital elevation models with aerial drones for environmental protection

A paper by Kazan Federal University was published in Drones.

In 2016, a group of researchers from the Department of Landscape Ecology of Kazan Federal University began developing a methodology for constructing high-resolution three-dimensional elevation models taken from an unmanned aerial vehicle survey. The methodology was developed in Tatarstan, which is mostly noted by flat landscapes. The group found that the accuracy of the model is comparable to the data obtained from ground-based scanning surveying instruments.

In the summer of 2017, an offer was made to the team to test the developed methodology in the mountainous terrain of the Sarykum section of the Dagestan Nature Reserve. At that time Sarykum was in preparations to open a geological park, and the survey would conveniently help to improve the existing methodology. The Sarykum barchan is one of the natural wonders of Dagestan. Interestingly, it is not located in a desert. Furthermore, it is the biggest dune not only in Russia but in the whole of Eurasia (and, apparently, the second biggest in the world after the Sahara dune), and its elevation, as the research showed, is 243 meters.

Because of the hilly relief, the survey of Sarykum was complicated by several factors. First, the area of 15 sq. km is quite large, and there had been no previous practice of surveying such an area with UAVs. Second, the complex mountainous terrain—the altitude deviation was 197 m—meant that the territory was difficult to survey from a single elevation. Third, difficult meteorological conditions—Sarykum is located in the valley of the Shura-Ozen River, in the foothills of the Narat-Tyube ridge, which provided unpredictable gusty winds.

A series of flight tasks consisting of 750*750 m squares from a height of 150 m above the takeoff point was developed for surveying the territory. Twenty six such squares were required to cover the entire area. For a precise positioning, a grid of 18 ground control points was used; their was determined by the Global Navigation Satellite Network (GNSS) receiver in real-time kinematic mode from the base station Kizilyurt, located within a 15 km range.

As a result of the survey, 6,400 images were received from June 10 to 16, 2017, which were processed and used for photogrammetric reconstruction of the area. The obtained coordinate point clouds were used to construct a digital elevation model; the spatial resolution of the final model was 0.5 m, which is more accurate than any model for this area. The accuracy of the model positioning according to the results of the error analysis is within the acceptable error limits according to GOST R 58854-2020 (Russian official standard). In addition to the high-resolution digital elevation model, an orthophoto plane with a spatial resolution of 10 cm was created.

Nowadays there are new, high-precision, and more productive unmanned aerial vehicles, which make such works much easier, especially on large territories. In the future, the researchers plan to perform works on surveying large territories using such systems, which will make it possible to decrease labor intensity manifold and to receive digital models and orthophoto maps with spatial positioning within 5 centimeters. However, such flying systems are very expensive, so many researchers continue to use budget UAVs in combination with ground control points.

The materials obtained in 2017 can be used for a geo-ecological assessment of the area, as well as for an automated assessment of overgrowth of the Sarykum barchan.

More information:
The Methodological Aspects of Constructing a High-Resolution DEM of Large Territories Using Low-Cost UAVs on the Example of the Sarycum Aeolian Complex, Dagestan, Russia

www.mdpi.com/2504-446X/5/1/7/htm

Provided by Kazan Federal University