No abstract

Abstract. Increasingly advanced and affordable close-range sensing techniques are employed by an ever-broadening range of users, with varying competence and experience. In this context a method was tested that uses photogrammetry and classification by machine learning to divide a point cloud into different surface type classes. The study site is a peat scarp 20 metres long in the actively eroding river bank of the Rotmoos valley near Obergurgl, Austria. Imagery from near-infra red (NIR) and conventional (RGB) sensors, georeferenced with coordinates of targets surveyed with a total station, was used to create a point cloud using structure from motion and dense image matching. NIR and RGB information were merged into a single point cloud and 18 geometric features were extracted using three different radii (0.02 m, 0.05 m and 0.1 m) totalling 58 variables on which to apply the machine learning classification. Segments representing six classes, dry grass, green grass, peat, rock, snow and target, were extracted from the point cloud and split into a training set and a testing set. A Random Forest machine learning model was trained using machine learning packages in the R-CRAN environment. The overall classification accuracy and Kappa Index were 98% and 97% respectively. Rock, snow and target classes had the highest producer and user accuracies. Dry and green grass had the highest omission (1.9% and 5.6% respectively) and commission errors (3.3% and 3.4% respectively). Analysis of feature importance revealed that the spectral descriptors (NIR, R, G, B) were by far the most important determinants followed by verticality at 0.1 m radius.

<div> <div>Recently, snow cover has gained a lot of interest as an important driver of plant species distribution in arctic and alpine environments, especially on small spatial scales . However, variation of snow cover at this scale is hardly resolved by open satellite data. Hence, linking remotely sensed snow cover and critical patterns and processes in vegetation can be challenging due to a mismatch in spatial resolution.</div> <div>We present a study based on a high alpine network of three webcams for the validation of snow cover products covering an entire year. Satellite based snow cover products (Landsat, Sentinel-2, downscaled MODIS products) are benchmarked on webcam-derived snow cover. While optical satellite remote sensing is a valuable tool for characterizing snow cover dynamics at the scale of tens of meters, cloud cover causes considerable data gaps. As a temporally and spatially more continuous estimate, we additionally produce meter-scale snow cover using the openAmundsen model, and we compare this to the webcam derived snow cover as well. For all datasets, ecologically relevant indicators like snow cover duration and the number of snow-free days are aggregated and validated both for the entire year and on a sub-seasonal scale.</div> </div>