**5. Conclusions**

This article proposed a UAS-based snow-field mapping technique and evaluated its performance at Mapa Republiky, a mountainous location of substantial scientific interest. The main benefit of the method rests in using ground control point-free UAS photogrammetry enabled by a custom-built multi-sensor system; this approach allowed us to perform an entire mapping mission from a safe distance, eliminating a major portion of the safety risks associated with avalanches. The parameters observed within the experiment, namely the georeferencing accuracy (approximately one decimeter RMSE) and spatial resolution of the photogrammetric products (centimeter-level), fully satisfied the requirements of the discussed application. Regrettably, the thick snow cover at the study site did not enable us to evaluate the SD mapping results achievable with a direct method, such as wire probing. In terms of the time consumption, our approach proved to be more efficient than the previously employed techniques, facilitating automatic execution of most processing steps. Despite this, there remains space for improvements, especially in view of the fact that some processing procedures currently depend on user intervention. Exploiting the outcomes presented herein, those related to safety in particular, the technique generally allows the user to collect data at Mapa Republiky very early in spring; such an operation was not feasible previously, meaning that the associated microclimate research at the given location had to face substantial objective limitations. This impediment can now be effectively overcome by using the novel approach.

**Author Contributions:** Conceptualization, P.G. and P.J.; Data curation, P.G.; Formal analysis, P.G. and P.J.; Funding acquisition, L.Z.; Investigation, P.G. and P.J.; Methodology, P.G. and P.J.; Project administration, L.Z.; Resources, P.G. and P.J.; Software, P.G.; Supervision, L.Z.; Validation, P.G., P.J. and J.H.; Visualization, P.G.; Writing—original draft, P.G.; Writing—review & editing, P.J., L.Z. and J.H.

**Funding:** This work was supported by the European Regional Development Fund under the project Robotics for Industry 4.0 (reg. No. CZ.02.1.01/0.0/0.0/15\_003/0000470).

**Conflicts of Interest:** The authors declare no conflict of interest.
