Philippos Garefalakis

Dr. Philippos Garefalakis

orcid

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
7 October 2022 in Earth Surface Dynamics
Verified
Journal Article
Open access
PDF file
DMDavid Mair
David Mair
APAriel Henrique Do Prado
Ariel Henrique Do Prado
PGPhilippos Garefalakis
Philippos Garefalakis
ALAlessandro Lechmann
Alessandro Lechmann
AWAlexander Whittaker
Alexander Whittaker
FSFritz Schlunegger
Fritz Schlunegger
Data on grain sizes of pebbles in gravel-bed rivers are of key importance for the understanding of river systems. To gather these data efficiently, low-cost UAV (uncrewed aerial vehicle) platforms have been used to collect images along rivers. Several methods to extract pebble size data from such UAV imagery have been proposed. Yet, despite the availability of information on the precision and accuracy of UAV surveys as well as knowledge of errors from image-based grain size measurements, open questions on how uncertainties influence the resulting grain size distributions still persist. Here we present the results of three close-range UAV surveys conducted along Swiss gravel-bed rivers with a consumer-grade UAV. We measure grain sizes on these images by segmenting grains, and we assess the dependency of the results and their uncertainties on the photogrammetric models. We employ a combined bootstrapping and Monte Carlo (MC) modeling approach to model percentile uncertainties while including uncertainty quantities from the photogrammetric model. Our results show that uncertainty in the grain size dataset is controlled by counting statistics, the selected processed image format, and the way the images are segmented. Therefore, our results highlight that grain size data are more precise and accurate, and largely independent of the quality of the photogrammetric model, if the data are extracted from single, undistorted nadir images in opposition to orthophoto mosaics. In addition, they reveal that environmental conditions (e.g., exposure to light), which control the quality of the photogrammetric model, also influence the detection of grains during image segmentation, which can lead to a higher uncertainty in the grain size dataset. Generally, these results indicate that even relatively imprecise and inaccurate UAV imagery can yield acceptable grain size data, under the conditions that the photogrammetric alignment was successful and that suitable image formats were selected (preferentially single, undistorted nadir images).

28 December 2023 in Earth Surface Processes and Landforms
Verified
Journal Article
Open access
DMDavid Mair
David Mair
GWGuillaume Witz
Guillaume Witz
APAriel Henrique Do Prado
Ariel Henrique Do Prado
PGPhilippos Garefalakis
Philippos Garefalakis
FSFritz Schlunegger
Fritz Schlunegger
The size of sedimentary particles in rivers bears information on the sediment entrainment or deposition mechanisms and the hydraulic conditions controlling them. However, collecting such data from coarse‐grained sediments is work intensive, both in the field and remotely. Therefore, attention has turned to machine learning models to improve the data acquisition. Despite their success, current methods need large quantities of data and yield results limited to a few percentile values of grain size datasets, often additionally affected by a systematic bias. In most cases, the root of these limitations is the challenge of accurately segmenting grains. Here, we present a new approach to improve the segmentation of individual grains based on the capacity of transfer learning in convolutional neural networks. Specifically, we re‐train a state‐of‐the‐art model for cell segmentation in biomedical images to find and segment coarse‐grained particles in images of fluvial sediments. Our results show that the performance in the segmentation tasks can be directly transferred to images of fluvial sediments and that our re‐trained models outperform existing methods. We document that our results are achievable with only 10%–20% of the data needed for training other deep learning models designed to measure the size of fluvial sediments. Moreover, we find that traits in our data control the segmentation performance. This enables data‐driven approaches to create specialist segmentation models. Additionally, comparing our automatically obtained datasets with the results retrieved from image and field‐based surveys confirms that improvements in segmentation are directly leading to more precise and more accurate grain size data even if data collection occurs in images taken at different conditions. Finally, we release a software package, the trained models and our data. The goal is to offer a tool to efficiently segment and measure grains in sediment images in an automated way, which can be adapted to different settings.

19 November 2019 in Solid Earth
Verified
Journal Article
Open access
PDF file
PGPhilippos Garefalakis
Philippos Garefalakis
FSFritz Schlunegger
Fritz Schlunegger
The stratigraphic architecture of the Swiss Molasse basin, situated on the northern side of the evolving Alps, reveals crucial information about the basin's geometry, its evolution, and the processes leading to the deposition of the siliciclastic sediments. Nevertheless, the formation of the Upper Marine Molasse (OMM) and the controls on the related Burdigalian transgression have still been a matter of scientific debate. During the time period from ca. 20 to 17 Ma, the Swiss Molasse basin was partly flooded by a shallow marine sea striking SW–NE. Previous studies have proposed that the transgression occurred in response to a rise in global sea level, a reduction of sediment flux, or an increase in tectonically controlled accommodation space. Here, we readdress this problem and extract stratigraphic signals from the Burdigalian molasse deposits that can be related to changes in sediment supply rate, variations in the eustatic sea level, and subduction tectonics. To achieve this goal, we conducted sedimentological and stratigraphic analyses of several sites across the entire Swiss Molasse basin. Field investigations show that the transgression and the subsequent evolution of the Burdigalian seaway was characterized by (i) a deepening and widening of the basin, (ii) phases of erosion and non-deposition during Lower Freshwater Molasse (USM), OMM, and Upper Freshwater Molasse (OSM) times, and (iii) changes in along-strike drainage reversals. We use these changes in the stratigraphic record to disentangle tectonic and surface controls on the facies evolution at various scales. As the most important mechanism, rollback subduction of the European mantle lithosphere most likely caused a further downwarping of the foreland plate, which we use to explain the deepening and widening of the Molasse basin, particularly at distal sites. In addition, subduction tectonics also caused the uplift of the Aar massif. This process was likely to have shifted the patterns of surface loads, thereby resulting in a buckling of the foreland plate and influencing the water depths in the basin. We use this mechanism to explain the establishment of distinct depositional settings, particularly the formation of subtidal shoals wherein a bulge in relation to this buckling is expected. The rise of the Aar massif also resulted in a reorganization of the drainage network in the Alpine hinterland, with the consequence that the sediment flux to the basin decreased. We consider this reduction in sediment supply to have amplified the tectonically controlled deepening of the Molasse basin. Because the marine conditions were generally very shallow, subtle changes in eustatic sea level contributed to the formation of several hiatuses that chronicle periods of erosion and non-sedimentation. These processes also amplified the tectonically induced increase in accommodation space during times of global sea level highstands. Whereas these mechanisms are capable of explaining the establishment of the Burdigalian seaway and the formation of distinct sedimentological niches in the Swiss Molasse basin, the drainage reversal during OMM times possibly requires a change in tectonic processes at the slab scale, most likely including the entire Alpine range between the Eastern and Central Alps. In conclusion, we consider rollback tectonics to be the main driving force controlling the transgression of the OMM in Switzerland, with contributions by the uplift of individual crustal blocks (here the Aar massif) and by a reduction of sediment supply. This reduction of sediment flux was likely to have been controlled by tectonic processes as well when basement blocks became uplifted, thereby modifying the catchment geometries. Eustatic changes in sea level explain the various hiatuses and amplified the deepening of the basin during eustatic highstand conditions.

7 March 2024 in Hydrology and Earth System Sciences
Verified
Journal Article
Open access
PDF file
APAriel Henrique Do Prado
Ariel Henrique Do Prado
DMDavid Mair
David Mair
PGPhilippos Garefalakis
Philippos Garefalakis
CSChantal Schmidt
Chantal Schmidt
AWAlexander Whittaker
Alexander Whittaker
FSFritz Schlunegger
Fritz Schlunegger
The construction of check dams is a common practice around the world where the aim is to reduce the damage by flooding events through mountain streams. However, quantifying the effectiveness of such engineering structures has remained very challenging and requires well-selected case studies, since the outcome of such an evaluation depends on site-specific geometric, geologic and climatic conditions. Conventionally, the check dams' effectiveness has been estimated using information about how the bedload sediment flux in the stream changes after the check dams are constructed. A permanent lowering of the bedload flux not only points to a success in reducing the probability of sediment transport occurrence but also implies that the sediment input through the system is likely to decrease. Here, we applied a method for data acquisition and two different equations (Meyer-Peter–Müller and Recking approach) to estimate and compare the sediment transport in a mountain stream in Switzerland under engineered and non-engineered conditions. Whereas the first equation is derived from a classical approach that is based on flume experiment data with a slope of less than 0.02 m m−1, the second equation (Recking) has been derived based on a bedload field dataset comprising active mountain streams under steeper conditions. We selected the Guerbe (Gürbe) River situated in the Swiss Alps as a case study, which has been engineered since the end of the 19th century. This has resulted in more than 110 check dams along a ca. 5 km reach where sediment has continuously been supplied from adjacent hillslopes, primarily by landsliding. We measured the riverbed grain size, topographic gradients and river widths within selected segments along this reach. Additionally, a gauging station downstream of the reach engineered with check dams yielded information to calibrate the hydroclimatic situation for the study reach, thus offering ideal conditions for our catchment-scale experiment. Using the acquired data and the dataset about historical runoff covering the time interval between 2009 and 2021 and considering the current engineered conditions, we estimated a mean annual volume of transported bedload which ranges from 900 to 6000 m3 yr−1. We then envisaged possible channel geometries before the check dams were constructed. We inferred (1) higher energy gradients which we averaged over the length of several check dams and which we considered a proxy for the steeper river slope under natural conditions; (2) channel widths that are smaller than those measured today, thereby anticipating that the channel was more confined in the past; and (3) larger grain size percentiles, which we consider to be similar to the values measured from preserved landslides in the region. Using such potential non-engineered scenarios as constraints, the two equations both point towards a larger sediment flux compared to the engineered state, although the results of these equations differed significantly in magnitude. Whereas the Recking approach returned estimates where the bedload sediment flux is ca. 10 times larger in comparison with the current situation, the use of the Meyer-Peter–Müller equation predicts an increase of ca. 100 times in bedload fluxes for a state without check dams. These results suggest that the check dams in the Guerbe River are highly efficient not only in regulating sediment transport by decreasing the probability of high sediment flux occurrence during torrential conditions but also in stabilizing the channel bed by avoiding incision. The most likely consequence is a stabilization of the terrain around such structures by reducing the activation of landslides.

9 January 2018 in Scientific Reports
Verified
Journal Article
Open access
PDF file
PGPhilippos Garefalakis
Philippos Garefalakis
FSFritz Schlunegger
Fritz Schlunegger
Large sediment fluxes from mountain belts have the potential to cause megafans to prograde into the neighbouring sedimentary basins. These mechanisms have been documented based from numerical modelling and stratigraphic records. However, little attention has been focused on inferring temporal changes in the concentrations of supplied sediment from coarse-grained deposits. Here, we extract changes of this variable in the field from a Late Oligocene, c. 4 km-thick suite of fluvial conglomerates situated in the North Alpine foreland basin, which evolved in response to the tectonic and erosional history of the Alps. We measure a decrease in channel depths from >2 m to <1 m and an increase in the largest grain sizes from 20 cm from the base to the top of the suite. These constraints are used to calculate an increase in fan surface slopes from 1.0° based on the Shields criteria for sediment entrainment. We combine slope and bulk grain size data with the Bagnold equation for sediment transport to infer higher concentrations of the supplied sediment. We use these shifts to propose a change towards faster erosion and a steeper landscape in the Alpine hinterland, driven by mantle-scale processes beneath the Alps.

31 August 2020 in Earth Surface Dynamics
Verified
Journal Article
Open access
PDF file
FSFritz Schlunegger
Fritz Schlunegger
RDRomain Delunel
Romain Delunel
PGPhilippos Garefalakis
Philippos Garefalakis
We present field observations from coarse-grained streams in the Swiss Alps and the Peruvian Andes to explore the controls on the probability of material entrainment. We calculate shear stress that is expected for a mean annual water discharge and compare these estimates with grain-specific critical shear stresses that we use as thresholds. We find that the probability of material transport largely depends on the sorting of the bed material, expressed by the D96∕D50 ratio, and the reach gradient but not on mean annual discharge. The results of regression analyses additionally suggest that among these variables, the sorting exerts the largest control on the transport probability of grains. Furthermore, because the sorting is significantly correlated neither to reach gradient nor to water discharge, we propose that the granulometric composition of the material represents an independent, yet important control on the motion of clasts in coarse-grained streams.