Search

Contemporary atmospheric temperature rise imposes profound physical changes to high- elevation mountain environments. These changes, by directly affecting the mountain cryosphere through glacier retreat and permafrost degradation, can alter the hydrologic regime of mountain drainage basins (Huggel et al., 2015), as well as the amount and caliber of sediment readily available for mass movements and fluvial transport. Although sediment supply exerts a primary control on channel stability and relevant geo-hazard potential, there is a general lack of standard procedures for characterizing sediment sources, sediment typology and for evaluating sediment availability. This gap, which is mainly associated with the variety of existing national and regional protocols of data collection, currently prevents pursuing an unbiased, transnational risk management strategy considering current climatic challenges. SedInOut, through a joint international effort, aims to develop methodologies for the quantification and characterization of sediment across representative pilot catchments, towards a sustainable land management that values geo-risk mitigation and sediment recycling. In this report, we present a methodological approach that relies on existing geological mapping (CARG project), high-resolution digital topography, and historical aerial photos and orthophoto mosaics, while integrating field-based and proximal sensing data in conjunction with multi- temporal, remotely based mapping (Figure 1).Remotely sensed procedures include multi- temporal mapping of glacier extent (Section 3.1.1), Quaternary materials (e.g., bedrock, glacial till, talus debris, colluvium, and alluvium) (Section 3.1.2), the drainage network (Section 3.1.3) and rapid shallow failures (i.e., sediment sources including debris slides, debris flows and bank collapses) (Section 3.1.4). Field-based and proximal sensing data include measurements on shallow landslide geometry (Section 3.2.1), and characterization of surface (i.e., manual Wolman pebble count and photo sieving) and subsurface (i.e., bulk sampling, on-site preliminary sieving, and laboratory sieving) grain size distribution (GSD) conducted at six representative sites (i.e., M1 through M6;Figure 2a) along the mountain channel network that drains the glaciated landscape of Upper Mazia Valley (Sections 3.2.2,3.2.3,3.3, and3.4). Field measurements on landslide geometry are critical for constraining an empirical landslide area-volume relation, which in turn is used for translating landslide areas, as mapped on sequential photo sets, into first-order volumetric estimates of mobilized debris. GSD data allows characterizing the spatial variability of characteristic sediment calibers (i.e., D50, D84, and D90) as well as the armoring ratio (an index of channel stability), starting at glacier and rock glacier fronts and moving downstream. The multi-temporal mapping approach is structured as follows. Through visual inspection of sequential orthophoto sets, we first track changes in glacier extent. Subsequently, as glaciers retreat, we map and quantify the extent of newly exposed Quaternary materials, the occurrence of shallow rapid failures, and the relevant changes in the structure of the main drainage network. In this document, we illustrate SedInOut methodological approach applied to Mazia Valley, here regarded as representative of conditions that characterize the Austroalpine geologic domain. In particular, we integrate two nested spatial scales: (i)Upper Mazia Valley(18.8 km2) over which we conducted extensive fieldwork to constrain the geometry of rapid shallow failures on the hillslopes and low-order streams, and to characterize alluvial sediment along the channel network(Figure 2a); (ii)Proglacial Mazia Area(8.4 km2), where we document decadal geomorphological changes following the retreat of the Mazia glacier (Figure 2b). Time and temporal scales of investigation are summarized inTable 1.

Il recente incremento della temperatura atmosferica è causa di profondi cambiamenti in ambienti montani di alta quota. Tali cambiamenti, influenzando direttamente la criosfera montana attraverso il ritiro dei ghiacciai e la degradazione del permafrost, possono alterare il regime idrologico dei bacini montani (Huggel et al., 2015), nonché la quantità e il calibro del sedimento disponibile per i movimenti di massa e il trasporto fluviale. Sebbene l'apporto di sedimento eserciti un controllo primario sulla stabilità dei canali e sul relativo potenziale rischio geologico, si nota una mancanza generale di procedure standard per la caratterizzazione di sorgenti e tipologia del sedimento, nonché per la valutazione della sua disponibilità. Questa lacuna, che è principalmente associata alla varietà dei protocolli nazionali e regionali esistenti per la raccolta dei dati, attualmente impedisce di perseguire una strategia di gestione del rischio oggettiva e transnazionale alla luce delle attuali sfide climatiche. SedInOut, attraverso uno sforzo interregionale congiunto, mira a sviluppare metodologie per la quantificazione e la caratterizzazione del sedimento in bacini idrografici rappresentativi, verso una gestione sostenibile del territorio che valorizzi la mitigazione del rischio geologico e l‘utilizzo del sedimento mobilizzato naturalmente. In questa relazione presentiamo un approccio metodologico che si basa sulla cartografia geologica esistente (progetto CARG), sulla topografia digitale ad alta risoluzione e sulle ortofotocarte storiche, integrando dati di rilevamento di campo e prossimali insieme a una mappatura multitemporale basata su metodi di telerilevamento (Figura 1). Le procedure di telerilevamento comprendono la mappatura multitemporale dell'estensione dei ghiacciai (sezione 3.1.1), dei depositi quaternari (ad es. roccia, till, detrito di versante, colluvium e alluvium) (sezione 3.1.2), del reticolo idrografico (sezione 3.1.3) e dei movimenti superficiali rapidi di versante, che per brevità, in questo documento verranno nominate come “frane rapide superficiali” (ad es. scivolamenti, colate di detrito e collassi spondali) (sezione 3.1.4). I dati di rilevamento sul campo e prossimali includono misurazioni sulla geometria delle frane poco profonde (sezione 3.2.1) e la caratterizzazione della granulometria superficiale (camminata casuale in stile Wolman e setacciatura basata su immagini) e subsuperficiale (ad esempio, campionamento ponderale, setacciatura meccanica preliminare in loco seguita da setacciatura in laboratorio) condotta in sei siti rappresentativi (da M1 a M6; Figura 2a) lungo il reticolo che drena il paesaggio glaciale dell'Alta Valle di Mazia (sezioni 3.2.2, 3.2.3, 3.3 e 3.4). Le misurazioni sul campo della geometria delle frane sono fondamentali per definire una relazione empirica area-volume, che a sua volta viene utilizzata per tradurre le aree di frana, mappate su set di foto sequenziali, in stime volumetriche del detrito mobilitato. I dati della distribuzione granulometrica consentono di caratterizzare la variabilità spaziale dei calibri di percentili caratteristici di una distribuzione granulometrica (D50, D84 e D90) e il rapporto di corazzamento (un indice di stabilità del canale), partendo dalle fronti dei ghiacciai e dei rock glacier, da cui i torrenti di studio si originano, e procedendo verso valle. L'approccio della mappatura multitemporale è strutturato come segue. Attraverso l'ispezione visiva di set di ortofotomosaici multitemporali, vengono tracciati i cambiamenti nell'estensione areale dei ghiacciai. Successivamente, con il ritiro dei ghiacciai, viene mappata e quantificata l'estensione dei materiali quaternari recentemente esposti e i relativi cambiamenti nella struttura del reticolo idrografico principale. In questo documento illustriamo l'approccio metodologico di SedInOut applicato alla Val di Mazia, considerata rappresentativa delle condizioni che caratterizzano il dominio geologico Austroalpino. In particolare, vengono integrate due scale spaziali: (i) l'Alta Val di Mazia (18,8 km2 ), in cui abbiamo condotto un'ampia attività di ricerca sul campo per definire la geometria delle frane rapide superficiali sui versanti e sui corsi d'acqua di basso ordine e per caratterizzare i sedimenti alluvionali lungo il reticolo idrografico (Figura 2a); (ii) l'Area Proglaciale di Mazia (8,4 km2 ), dove abbiamo documentato i cambiamenti geomorfologici decennali in seguito al ritiro del Ghiacciaio di Mazia (Figura 2b). Le scale temporali e spaziali di indagine sono riassunte in Tabella 1., La ricerca è stata finanziata dal Progetto Interreg V-A Italy – Austria ITAT3032 SedInOut – 'Development of a risk management methodology through the assessment of sediment availability for mass wasting in a mountain environment'

Channel morphology of forested, mountain streams in glaciated landscapes is regulated by a complex suite of processes, and remains difficult to predict. Here, we analyze models of channel geometry against a comprehensive field dataset collected in two previously glaciated basins in Haida Gwaii, B.C., to explore the influence of variable hillslope–channel coupling imposed by the glacial legacy on channel form. Our objective is to better understand the relation between hillslope–channel coupling and stream character within glaciated basins. We find that the glacial legacy on landscape structure is characterized by relatively large spatial variation in hillslope–channel coupling. Spatial differences in coupling influence the frequency and magnitude of coarse sediment and woody material delivery to the channel network. Analyses using a model for channel gradient and multiple models for width and depth show that hillslope–channel coupling and high wood loading induce deviations from standard downstream predictions for all three variables in the study basins. Examination of model residuals using Boosted Regression Trees and nine additional channel variables indicates that ~10 to ~40% of residual variance can be explained by logjam variables, ~15–40% by the degree of hillslope–channel coupling, and 10–20% by proximity to slope failures. These results indicate that channel classification systems incorporating hillslope–channel coupling, and, indirectly, the catchment glacial legacy, may present a more complete understanding of mountain channels. From these results, we propose a conceptual framework which describes the linkages between landscape history, hillslope–channel coupling, and channel form. © 2018 John Wiley & Sons, Ltd.

_, Geomorphology for Society, Conference proceedings

Climate change, human activity and other environmental disturbances can affect Earth surface systems via alteration of vegetation cover, hydrologic regime, permafrost distribution, and glacier fluctuations. For example, Pleistocene temperature drop had profound climatic and geomorphic influences worldwide, including abrupt glacial advance in vast terrestrial areas that have both sculpted the topography and deposited extensive mantles of glacigenic materials. In consideration of such legacies, formerly glaciated landscapes today can be considered at a unique stage of readjustment (recovery) with respect to the spatial organization of currently active geomorphic process domains (Brardinoni and Hassan, 2006), and the magnitude and patterns of sediment fluxes (Church and Slaymaker, 1989; Hewitt et al., 2002;Warburton, 2007). Accordingly, both contemporary environmental changes and disturbances over the Quaternary are significantly influencing patterns of weathering, detachment, transport, and deposition of material across landscape components. It is a challenge to develop a better understanding of how such changes and disturbances interact to modulate sedimentary source-to-sink fluxes and budgets in the light of peculiar landscape sensitivities (Slaymaker et al., 2009).

We monitor bedload transport and water discharge at six stations in two forested headwater streams of the Columbia Mountains, Canada. The nested monitoring network is designed to examine the effects of channel bed texture, and the influence of alluvial (i.e.step pools and riffle pools) and semialluvial morphologies (i.e.boulder cascades and forced step pools) on bedload entrainment and transport. Results indicate that dynamics of bedload entrainment are influenced by differences in flow resistance attributable to morphology. Scaled fractional analysis shows that in reaches with high form resistance most bedload transport occurs in partial mobility fashion relative to the available bed material, while calibers finer than 16mm attain full mobility during bankfull flows. Equal mobility transport for a wider range of grain sizes is achieved in reaches exhibiting reduced form resistance. Our findings confirm that the Shields value for mobilization of the median surface grain size depends on channel gradient and relative submergence; however, we also find that these relations vary considerably for cobble and gravel bed channels due to proportionality between dimensionless shear stress and grain size. Exponents of bedload rating curves across sites correlate most with the D90s of the mobile bed, however, where grain effects are controlled (i.e.along individual streams), differences in form resistance across morphologies exert a primary control on bedload transport dynamics. Application of empirical formulae developed for use in steep alpine channels present variable success in predicting transport rates in forested snowmelt streams. Formulae that explicitly account for reductions in mobile bed area and high morphological resistance associated with woody debris provide the best approximation to observed empirical data.

To evaluate landslide response of mountain drainage basins to forest management we examine a historical inventory in northeastern Vancouver Island. The study area is underlain by extrusive (upper Triassic Karmutsen Formation) and intrusive rocks (Jurassic Island Intrusions). The dataset, compiled via interpretation of sequential aerial photosets, helicopter traverses, and extensive fieldwork, comprises a total of 1961 sediment sources, including 798 field-measured events. Field- and photo-based data cover a time window of approximately 70 years. The effects of forest management on landslide activity are assessed in terms of landslide density, sediment production, landslide geometry, landslide magnitude-frequency relations, topographic conditions of landslide initiation and deposition, and sediment redistribution across landscape components. Results indicate that forest management alters natural landslide dynamics in many respects. Logging-related debris avalanches are typically smaller. Consequently, the magnitude-frequency relation in logged terrain occupies the small-medium magnitude spectrum (< 6000 m3), with frequencies increasing by over an order of magnitude. Lithologic effects on sediment production appear amplified, in that terrain underlain by extrusive rocks become increasingly more unstable than intrusive ones. Analysis of landslide initiation and deposition zones reveals that forest management promotes colluvial aggradation on mid and lower hillslopes and in gullies. This pattern, which accelerates sediment recharge of gully systems, has the potential of increasing the frequency of channelized debris-flows, hence cause an extended period of disturbance, before sediment dynamics recover to pre-logging conditions. The effects of forest clearing on hillslope-channel coupling are composite: in cutblocks the percentage of sediment delivered to streams is reduced by 20-60% whereas road-related landslides are associated with highestconnectivity to streams.

Basin-wide erosion rates can be determined through the analysis of in situ-produced cosmogenic nuclides based on the idea that samples taken at the outlet of a catchment are representative of the entire upstream basin. However, this is only true when a certain number of assumptions are verified. Among the most important assumptions is that denudation rates are uniform through space and time; i.e. the catchment is in cosmogenic steady-state. If this is the case, then the in-going nuclide flux through production is equal to the out-going flux through decay and erosion, and the system is in isotopic equilibrium. When this condition is not matched, the calculation of basin-wide erosion rates through cosmogenic nuclides is subject to the well-mixing conditions of the alluvial sediment in the rivers. Here, we bring a new contribution towards the understanding of sediment mixing effects on in-situ cosmogenic nuclide concentrations in alluvial sediment. We proceed by following the methodology proposed by Binnie et al. (2006), and present the results from a sediment mixing model for a small ( 42 km2) catchment in the Alps. The model results show the importance of the sampling location and of the catchment’s size to have reliable erosion rates. However, the models also highlight the importance of knowing the sediment transfer mechanisms and the processes magnitude. When the transport of sediment mainly occurs through soil creep and shallow landslides, high concentration material is supplied into the fluvial system, increasing the 10Be concentration in the alluvial sediment. Likewise, the high-frequency of mass-wasting processes or the occurrence of sporadic but large-magnitude events, allow the supply of low-concentration sediment that result in a lower cosmogenic nuclide signal in the channels. A strong bias arises for the calculation of basin-wide erosion rates if mass-wasting processes dominate the mechanisms of sediment transfer in the catchment. In this situation the use of cosmogenic nuclides signal results in an over- or underestimation of the erosion rates. Therefore, our results suggest that when using cosmogenic nuclides, the achievement of mixing conditions in alluvial sediment is of primary importance for the correct estimation of the erosion rates and that this latter has to be accurately interpreted when the mixing conditions are unknown or not achieved.

To characterize channel-network morphodynamics and response potential in the Gadria-Strimm basin (14.8 km^2) we conduct a concerted effort entailing: (i) high-resolution mapping of landforms, channel reaches, and sediment sources; and (ii) historical evolution of colluvial channel disturbance through sequential aerial photosets (1945-59-69-82-90-00-06-11). The mapping was carried out via stereographic inspection of aerial photographs, examination of 2.5-m gridded DTM and DSM, and extensive field work. The study area is a formerly glaciated basin characterized by peculiar landform assemblages imposed by a combination of tectonic and glacial first-order structures. The most striking feature in Strimm Creek is a structurally-controlled valley step separating an upper hanging valley, dominated by periglacial and fluvial processes, and a V-notched lower part in which lateral colluvial channels are directly connected to Strimm’s main stem. In Gadria Creek, massive kame terraces located in proximity of the headwaters provide virtually unlimited sediment supply to frequent debris-flow activity, making this sub-catchment an ideal site for monitoring, hence studying the mechanics of these processes. Preliminary results point to a high spatial variability of the colluvial channel network, in which sub-sectors have remained consistently active during the study period while others have become progressively dormant with notable forest re-growth. In an attempt to link sediment flux to topography and substrate type, future work will involve photogrammetric analysis across the sequential aerial photosets as well as a morphometric/geomechanical characterization of the surficial materials.

Debris flows are efficient processes of sediment transfer from slope base to alluvial fans in mountain drainage basins. To advance current understanding of debris-flow sediment dynamics at the regional scale we analyze a historical (1998-2009) database of debris flows in 82 selected basins of the Alto Adige Province, north-eastern Italy. By combining historical information on event-based debris-flow volumetric deposition, high-resolution (LiDAR-derived) digital topography, and mapping of Quaternary sediment stores we are able to (i) characterize the principal topographic conditions at which debris-flow initiation, transportation, and deposition operate; (ii) detail debris-flow sedimentary signatures across spatial scales; and (iii) link debris-flow sediment flux to morphometry, lithologic variability, and sediment availability. We show that basin-wide specific yield obeys a negative scaling relation with basin area, which in turn is strongly controlled by dominant bedrock geology and abundance of Quaternary deposits. When the event-based sediment flux is subsumed across spatial scales, a more complex sedimentary signature is apparent, regardless of sediment availability. We hypothesize that this signature is characteristic of most debris-flow dominated settings.

Debris flows are one of the most relevant natural hazards in mountain regions. A combination of structural (e.g. check-dams) and non structural (e.g. warning systems) measures is needed in most cases to reduce debris flow risks, and both benefit from experimental data on debris-flow characteristics. A new experimental site is being equipped in the Autonomous Province of Bolzano (Italy) for both monitoring purposes and testing warning systems. The study site (Gadria basin) is a small channel subjected to frequent debris flows. The overall system includes the following components: 4 rain gauges located at different elevations, 5 radar sensors and 5 geophones, and 3 video cameras and flash lights to record the propagation and the deposition of the debris flow. Transmission of data and alerts from the instruments will use radio technology because GSM coverage is not available in the basin.