Comparing UAS LiDAR and Structure-from-Motion Photogrammetry for Peatland Mapping and Virtual Reality (VR) Visualization

Abstract

The mapping of peatland microtopography (e.g., hummocks and hollows) is key for understanding and modeling complex hydrological and biochemical processes. Here we compare unmanned aerial system (UAS) derived structure-from-motion (SfM) photogrammetry and LiDAR point clouds and digital surface models of an ombrotrophic bog, and we assess the utility of these technologies in terms of payload, efficiency, and end product quality (e.g., point density, microform representation, etc.). In addition, given their generally poor accessibility and fragility, peatlands provide an ideal model to test the usability of virtual reality (VR) and augmented reality (AR) visualizations. As an integrated system, the LiDAR implementation was found to be more straightforward, with fewer points of potential failure (e.g., hardware interactions). It was also more efficient for data collection (10 vs. 18 minutes for 1.17 ha) and produced considerably smaller file sizes (e.g., 51 MB vs. 1 GB). However, SfM provided higher spatial detail of the microforms due to its greater point density (570.4 vs. 19.4 pts/m2). Our VR/AR assessment revealed that the most immersive user experience was achieved from the Oculus Quest 2 compared to Google Cardboard VR viewers or mobile AR, showcasing the potential of VR for natural sciences in different environments. We expect VR implementations in environmental sciences to become more popular, as evaluations such as the one shown in our study are carried out for different ecosystems.

Unmanned Aircraft System (UAS) Structure-From-Motion (SfM) for Monitoring the Changed Flow Paths and Wetness in Minerotrophic Peatland Restoration

“…Correspondingly, the volume of the peat varies according to its water content [87]. Simultaneously, the 3D structure of the interpreted peatland surface altered according to the developing phenology of the annual understory vegetation (Figure 9A1-A5, [22]). For instance, the middle part of Loukkusuo is dominated by grasses that might have increased the surface elevation during the observation period from June to August.…”

“…A major issue with the SfM in vegetated environments is its tendency to model the ground surface into the canopy of any prevailing dense understory vegetation, such as sedges or dwarf shrubs (Figure 9A1-A5), whereas LiDAR technology typically measures points through the vegetation, coming closer to the so-called surface of the vegetated peatland (Sphagnum canopy in many cases) [22]. However, LiDAR accuracies also decrease under dense vegetation [36].…”

“…However, the flow paths can also be simulated by analyzing a digital elevation model (DEM) (e.g., [21]). The highly detailed digital surface model (DSM) and its ground-filtered derivative, digital terrain model (DTM), can be produced using airborne remote sensing [22]. Unmanned aircraft systems (UAS), in particular, allow photogrammetric mapping at a centimetre-level spatial resolution [23].…”

“…Photogrammetric SfM can reach a spatial accuracy comparable with laser-based light detection and ranging (LiDAR), except in the case of densely vegetated areas where the land surface remains mostly obscured for the cameras but not for all laser beams [18]. UAS-derived DSMs and DTMs have been increasingly used to study the morphometry of peatland surfaces (e.g., [18,22,27]. Several UAS methods have also been applied in peatland restoration monitoring.…”

Before and After: A Multiscale Remote Sensing Assessment of the Sinop Dam, Mato Grosso, Brazil

“…VR not only increases the understanding of place, but can also help develop fieldwork observational techniques, while developing data interpretation skills of increasing importance in the future [116]. Virtual reality allows users to interact with the models and not only gain a more real sense of the study site, but also raise broader interest in the project [117].…”