In this thesis, we discuss data on the total ascent along 250 recorded GPS tracks, situated in western Slovenia. GPS tracks were recorded with a simple single-frequency code GPS recorder Qstarz BT-Q1000P with distance criteria of 10 m from the previously recorded point. From all recorded GPS tracks, we calculated the total ascent in different ways, namely directly from the recorded ellipsoidal heights, using different thresholds of height profile simplification, with the help of Google Earth in view mode of the recorded GPS track “Absolute” or “Clamped to ground”, meaning on the SRTM 90 m, and by laying the GPS tracks on DMV 1 m data from a recent recording of Slovenia with ALS, which we used as a reference. We evaluated the suitability of a particular method of etermining the total ascent and empirically determined correction factors to ensure comparability of data with reference values and their mutual comparability. Based on the aggregated processing of 250 recorded GPS tracks, we found out that the total ascent, calculated from originally recorded, uncorrected GPS tracks, has to be reduced for 20 % on average, to get a value that is comparable to the reference. Total ascents, obtained with Google Earth in view mode “Clamped to ground” or calculated by the simplification threshold of 15 m, are on average too small due to more generalized terrain model or height profile, therefore it is necessary to increase them for 20 % on average to ensure the comparison with the reference values. It turns out that the sum of deviations, calculated by comparing each total ascent to its reference value, is on average the smallest when total ascent is etermined using 5 m threshold of height profile simplification.
|
