Abstract
A comparison between errors associated with snow-cover reconstruction performed by processing aerial imagery acquired by a visible-light camera mounted on board unmanned aerial vehicles, one the one hand; and average terrain roughness, on the other, revealed a dependent relationship between these variables. A stronger correlation is noted for two of the studied test areas (Polana Izerska and Krobica, both located in SW Poland), as opposed to the remaining site (Drożyna, SW Poland). In particular, correlations are noticeable where the analysis is performed in moving windows. It is typical for terrain where depth of snow cover is reconstructed with severe errors to reveal a high degree of roughness caused by single trees, clumps of trees or buildings. Ambiguous results are obtained for the Drożyna research field. While the character of the dependent relationship there seems consistent with results for the remaining sites, the strength is low. The lower values for the correlation coefficient were driven by observations for which errors were found to be high while values for the Topographic Ruggedness Index were at the same time low. This effect can be explained by reference to the specific nature of the area reconstructed, which is much transformed by human activity. It proves difficult to reconstruct the depth of snow cover on roads properly, as these are either partially cleared or snow or characterised by its loss in the course of melting. Low thickness of snow cover is thus found to be a constrained when it comes to the generation of accurate reconstructions of the depth of snow cover. This is in fact a finding in agreement with what has been reported by other authors.
Highlights
To be high while values for the Topographic Ruggedness Index were at the same time low
A comparison between errors associated with snow-cover reconstruction performed by processing aerial imagery acquired by a visible-light camera mounted on board unmanned aerial vehicles, one the one hand; and average terrain roughness, on the other, revealed a dependent relationship between these variables
A stronger correlation is noted for two of the studied test areas (Polana Izerska and Krobica, both located in SW Poland), as opposed to the remaining site (Drożyna, SW Poland)
Summary
Obszar badań obejmował trzy pola badawcze: Drożyna, Krobica i Polana Izerska. Pod względem administracyjnym znajdowały się one na obszarze województwa dolnośląskiego. Pozyskane warstwy grubości pokrywy śnieżnej posiadały rozdzielczość 1 m, co dało odpowiednio: Krobica – 426x240 px, Drożyna – 201x201 px, Polana Izerska – 425x276 px. Przez SfM rozumie się zbiór połączonych ze sobą algorytmów, których celem jest utworzenie trójwymiarowego numerycznego modelu powierzchni terenu, na podstawie zbioru zdjęć danego obiektu lub grupy obiektów, wykonanych pod różnym kątem. Zarówno warstwa zasięgu wyznaczanego metodą ISO, jak i warstwa zasięgu pokrywy śnieżnej bazująca na jej grubości obliczonej algorytmem SfM, poddane zostały dodatkowemu przetworzeniu wstępnemu, polegającemu na usunięciu pojedynczych pikseli, które w swoim sąsiedztwie otoczone były przez co najmniej 5 pikseli o innej wartości niż one same. Ujednolicone rastry zostały porównane, stosując różnicę zasięgu pokrywy śnieżnej (dane referencyjne bazujące na metodzie ISO) i zasięgu otrzymanego poprzez reklasyfikację warstwy grubości śniegu (w oparciu o SfM). Możliwe kombinacje wartości dla warstwy oraz ich interpretacja (SE – zasięg pokrywy śnieżnej warstwy referencyjnej, SE-SD – modelowany zasięg pokrywy śnieżnej otrzymany przez reklasyfikację warstwy grubości pokrywy śnieżnej) Possible combinations of values for a difference layer and their interpretation (SE – extent of snow cover in the reference layer, SE-SD – modelled extent of snow cover determined through the reclassification of thickness of the layer of snow)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
