Search

Rapid mass movements are a major threat in populated landscapes, as they can cause significant loss of life and damage civil infrastructure. Previous work has shown that using environmental seismology methods to monitor such mass movements and establish monitoring systems offers advantages over existing approaches. The first important step in developing an early warning system for rapid mass movements based on seismic signals is automatically detecting events of interest. Though the approach, such as short-term average to long-term average ratio (STA/LTA) and machine learning model, was introduced to detect events (e.g., debris flow and rockfall), it is still challenging to calibrate input parameters and migrate existing methods to other catchments. Detection of debris flows, for instance, is similar to anomaly detection if we consider the seismic stations recording background signals as an overwhelming majority condition. Benford's law describes the probability distribution of the first non-zero digits in numerical datasets, which provides a functional, computationally cheap approach to anomaly detection, such as fraud detection in financial data or earthquake detection in seismic signals. In this study, seismic signals generated by rapid mass movements were collected to check the agreement of the distribution of the first digit with Benford's law. Subsequently, we develop a computationally efficient and non-site-specific model to detect events based on Benford's law using debris flows from the Illgraben, a Swiss torrent, as an example. Our results show that seismic signals generated by high-energy mass movements, such as debris flows, landslides, and lahars, follow Benford's law, while those generated by rockfall and background signals do not. Furthermore, our detector performance in picking debris-flow events is comparable to a published random forest and seismic network-based approach. Our method can be applied at other sites to detect debris-flow events without additional calibration and offers the potential for real-time warnings.

In the coming decades with enhanced rainstorm activity, massive sediment redistribution in Alpine catchments will be a key hazard and challenge in Alpine communities. While several studies have collected data from massive rock slope failures, few studies have quantitatively assessed the cascading sediment redistribution in highly active alpine catchments. Recurrence intervals for cliffs falls are estimated at 80 years (Krautblatter et al., 2012 ), thus, observations of the subsequent sediment cascading are limited or inexistent despite their major role in landscape evolution and sediment fluxes. Digital aerial photogrammetry acquired by governmental agencies is becoming a relevant tool to better understand short landscape response to climate change. Repetitive yearly to bi-yearly orthophotos and DSM extracted from large format aerial surveys represent a valuable monitoring tool at regional scale because of their wide extent coverage (km) at a high spatial resolution (20 cm). This contribution reports the massive sediment redistribution that has been triggered by the multistage failure of >200.000 m³ from the Hochvogel dolomite peak during the summer of 2016. Seven true orthophotos and high-resolution aerial photogrammetric digital surface models (DSM) between 2010 and 2020 were 3D coregistered to a reference system for optimized volume calculation in steep terrain. Three consecutive differential DSMs (2010-2012, 2012-2014, 2014-2015) describe the catchment morphodynamics before the cliff fall, while, the subsequent differential DSMs (2015-2017, 2017-2018, 2018-2020) describe the morphodynamics one year, two years and four years after the cliff fall. Spectrograms from surrounding seismic stations expand the understanding of the cliff fall timing. We observe the decadal throughput of >200.000 m³ of sediment with massive sediments pulses that (i) respond with reaction times of 0-4 years and relaxation times beyond 10 years, (ii) with faster 0-2 years response times in the upper catchment (A&B) and >>2 years response times in the lower catchments, (iii) the inversion of sedimentary (>10²-10³ mm/a) to massive erosive regimes (>10² mm/a) within single years and vice versa and the (iv) dependency of redistribution to rainstorm frequency and intensities. Krautblatter, M., Moser, M., Schrott, L., Wolf, J., Morche, D., 2012. Significance of rockfall magnitude and carbonate dissolution for rock slope erosion and geomorphic work on Alpine limestone cliffs (Reintal, German Alps). Geomorphology 167, 21–34. https://doi.org/10.1016/j.geomorph.2012.04.007

Weather conditions are an important driver of Earth surface dynamics, such as gravitational mass wasting, flood propagation, biological activity events and physical interactions within the critical zone. While there are dedicated sensors to capture meteorological parameters, these sensors are comparably expensive, have a small spatial footprint and often lack the temporal resolution needed to constrain high frequency meteorological dynamics. We introduce the concept of meteo-seismology, i.e. the measurement of first-order ground motion signatures of weather conditions by decisively installed seismic sensors. While meteorological manifestations are generally considered seismic noise and it may seem odd to use seismometers instead of weather stations, geophysical sensors circumvent or complement the above caveats and add further important data to a comprehensive picture of the rapidly changing state of the atmosphere and its interaction with the landscape we live in. Based on examples from prototype forested landscapes in Central Europe and Chilean Patagonia, we demonstrate how seismic stations can be used to infer properties of the pressure and wind field and its coupling to the biosphere, constrain rain intensity and drop properties, yield temperature proxies and their propagation into the ground, and survey ground moisture trends and discharge patterns. Understanding the seismic signatures of a meteorological origin also allows to, vice versa, better handle the contaminating side of these seismic sources in records, where high frequency signals are to be used for other than meteo-seismological studies. Our approach offers an alternative and complementary way to non-invasively monitor hydrometeorological energy and matter fluxes at high temporal and spatial resolution.

On 14th and 15th July 2021 heavy rainfall in western Germany, Belgium and the Netherlands caused severe floodings. The most affected area in Germany was the 86 km long Ahr river valley, which suffered from severe damage to buildings and infrastructure and where more than 130 people died. The Ahr flood exceeded a return period of at minimum 500 years. The river Ahr drains around 900 km2 of the Rhenish Massif with a dendritic catchment from west to east causing differences in slope properties and covering different land uses. The flood water carried large woody debris that caused clogging in bridges of the main valley and some tributaries, some of which collapsed. This extreme event thus offers the opportunity to explore the spatial impact and characteristics of large wood on channel dynamics. This study aims to find thresholds for the initiation of large wood recruitment, dependent on catchment size, valley slopes, water quantity and land use.The study focuses on the whole catchment area of the Ahr river. Using general vegetation data obtained from the German national forest inventory, we quantified the type and amount of flood-affected vegetation. We adopted an NDVI (normalized differential vegetation index) based change detection approach using Landsat/Sentinel satellite data (Google Earth Engine based Hazmapper) to identify recruited live vegetation and deadwood transport during the flood. We validated this remotely obtained data with field surveys along selected valley sections.Large wood was predominantly recruited from the fluvial corridor in the main Ahr valley and not from the tributaries, even if those experienced heavy precipitation and surface runoff (up to few meters high discharge on the flood plains). Although we have observed transported tree trunks in those tributaries, there was no large pattern. We aspire to identify deposition areas using ortho photos to investigate a wood balance.Including large wood in flood modelling would improve flood hazard assessments. Remote sensing analyses offer an interim solution in this regard by helping to identify potential large wood recruitment areas and inform designing flood hazard prevention measures.

Alpine glaciers have been rapidly retreating and at increasing rates in recent decades due to climate warming. As a consequence, large amounts of suspended- and bed-load flux are being released to proglacial environments, such as proglacial forefields. These regions are among the most unstable geomorphic systems of the Earth because they rapidly respond to changing discharge and sediment conditions. Given this, it might be hypothesized that their intense morphodynamic activity, being a complex and non-linear process, could “shred” the sediment transport signal itself, and especially that related to subglacial sediment export.To date, our knowledge on subglacial sediment export by subglacial streams is essentially dominated by suspended sediment dynamics recorded in front of shrinking glaciers because of the limitations in measuring bedload transport. The latter is usually monitored far downstream from glacier termini by permanent stations (e.g. water intakes, geophone systems) leaving major uncertainties in the absolute amounts and temporal patterns of transport in both glacial and proglacial environments, as well as the relative importance compared to suspended sediment in case of morphodynamic filtering. Thus, the aim of this project was to investigate the evolution of the both suspended- and bedload subglacial export signals within the proglacial forefield to quantify the extent and the timescale over which proglacial morphodynamics filter them.This work focuses on a large Alpine glacial forefield, almost 2 km in length, that has formed since the early 1980s at the Glacier d’Otemma (southern-western Swiss Alps, Valais). Data were collected over two entire melt seasons (June-September 2020 and 2021) experiencing different climatic conditions, the first year warm and relatively dry and the second cold and relatively wet. Suspended transport was recorded using conventional turbidity-suspended sediment concentration relationship, bedload transport was monitored seismically, while the morphodynamic filtering was determined using signal post-processing techniques. At present, there are no studies combining continuous measurements of both suspended- and bed-loads in such environments.Results show that the signal of subglacial bedload export, unlike suspended load export, is rapidly shredded by proglacial stream morphodynamics, which we show is due to a particle-size dependent autogenic sorting of sediment transport at both daily and seasonal time-scales. The result is that over very short distances, the signal of subglacial bedload sediment export is lost and replaced by a signal dominated by morphodynamic reworking of the proglacial braidplain. The suspended signal is less impeded but significant floodplain storage and release of suspended sediment was observed. These results question the reliability of current inferences of glacial erosion rates from sediment transport rates often measured some way downstream of glacier margins.

Rock slope instabilities cause significant risk in populated alpine areas. To anticipate the final failure, a detailed understanding of the preparatory process dynamics including all potential promoting and triggering factors is needed. While standard external and internal drivers are known, measured evidence and a quantification of their relevance at a specific site is often lacking.Here, we present the evolution of the imminent Hochvogel summit failure (200,000–600,000 m³) over multiple decades towards the current highly sensitive system. We identified the three most relevant potential drivers at the Hochvogel instability: (i) earthquakes, (ii) seasonal and short-term meteorological effects and (iii) increasing internal stress. To quantify these, we use diverse sources of information. Earthquake catalogues and the records of the regional seismic broadband stations help to constrain known historical rock fall events at the Hochvogel. The effect of precipitation events, snowmelt and temperature is quantified by the analysis of high-resolution crack opening and rain data of the last four years. Finally, we exploit the record of our local seismic network to reveal internal rock bridge failures, rock fall activity in the flanks and the seismic stressing of the instable mass due to local earthquakes.The current process dynamics prove a close-to-failure status of the instability. The combination of historic records and high-resolution real-time data not only makes the Hochvogel a benchmark site for alpine hazard early warning but also enables the comprehensive definition and quantification of its relevant drivers. This will improve the global understanding of rock failure dynamics and so the anticipation ability for instable rock slopes.

Large Alpine glaciers typically export significant quantities of subglacially eroded sediment to their downstream environments via meltwater streams. The supply of this glaciogenically-produced sediment has an extensive impact on landform genesis and evolution; downstream ecology and ecosystem succession; river morphology and flood risk management; hydropower installations and industrial sediment extraction. However, little is known about the processes and mechanisms driving the entrainment and evacuation of subglacial sediment by meltwater flow. Controls on the rate of sediment export remain poorly constrained, limiting the reliability of future sediment yield predictions. Although the fine (suspended) fraction of meltwater-borne subglacial sediment export is relatively well studied, records of suspended sediment load are commonly obtained from distal proglacial locations and may consequently be affected by filtering effects in proglacial rivers and forefields. Additionally, almost nothing is known about the coarse (bedload) component and its relationship to suspended sediment transport due to the difficulties in monitoring bedload flux. In this study, we present a first-of-its-kind dataset from Glacier d’Otemma, Switzerland, capturing three years (2020, 2021 and 2022) of continuous suspended sediment load, bedload and water discharge data from a monitoring site immediately in front of the glacier meltwater portal. Measurements of coarse sediment flux were performed using recently developed environmental seismological monitoring techniques. Data from all three years show that bedload delivery to the glacier margin switches from being transport-limited to being supply-limited during the melt season and was driven by snow-line retreat. Bedload evacuation reduces to very low levels as soon as the snow-line reaches the upper parts of the glacier, which it does increasingly often due to climate warming. We hypothesise that the position of the snow line controls the degree of channelisation – or the efficiency – of the subglacial drainage network above and below it, with consequences for both sediment availability and channel competence. Meltwater in the non-channelised parts of the glacier struggles to transport bedload. As the channelisation develops, accumulated bedload is evacuated until the point at which the subglacial channel network reaches its maximum extent or, due to progressive channel division in a dendritic network, is developing channels that are too small to readily transport it. Thus, recent reports of increases in bedload yield from Alpine glaciers may not reflect increases in glacial erosion but simply an increase in the upstream elevation of snow-lines, the associated subglacial channelization and hence subglacial bedload export capacity.

Seafloor sediment flows (turbidity currents) form some of the largest sediment accumulations on Earth, carry globally significant volumes of organic carbon, and can damage critical seafloor infrastructure. These fast and destructive events are notoriously challenging to measure in action, as they often damage any instruments anchored within the flow. We present the first direct evidence that turbidity currents generate seismic signals which can be remotely sensed (~1-3 km away), revealing the internal structure and remarkably prolonged duration of the longest runout sediment flows on Earth. Passive Ocean Bottom Seismograph (OBS) sensors, located on terraces of the Congo Canyon, offshore West Africa, recorded thirteen turbidity currents over an 8-month period. The occurrence and timing of these turbidity currents was confirmed by nearby moorings with acoustic Doppler current profilers.Results show that turbidity currents travelling over ~1.5 m/s produce a seismic signal concentrated below 10 Hz with a sudden onset and more gentle decay. Comparison of the seismic signals with information on flow velocities from the acoustic Doppler current profilers demonstrates that the seismic signal is generated by the fast-moving front of the flow (frontal cell), which contains higher sediment concentrations compared to the slower-moving body. Long runout flows travelling >1000 km have a fast (3.7-7.6 m s-1) frontal cell, which can be 14 hours, and ~350 km long, with individual flows lasting >3 weeks. Flows travelling >1000 km eroded >1300 Mt of sediment in one year, yet had near-constant front speeds, contrary to past theory. The seismic dataset allows us to propose a fundamental new model for how turbidity currents self-sustain, where sediment fluxes into and from a dense frontal layer are near-balanced.Seismic monitoring of turbidity currents provides a new method to record these hazardous submarine flows, safely, over large areas, continuously for years yet at sub-second temporal resolution. Monitoring these processes from land would considerably ease deployment efforts and costs. Thus, work is underway investigating if terrestrial seismic stations can record submarine seafloor processes in Bute Inlet, a fjord in western Canada where independent measurement of delta-lip failures and turbidity currents can be compared to a passive seismic dataset.

Climate change affects temperate forests particularly by changes in water availability as a result of rising temperatures and changing precipitation dynamics. While the annual mean will remain roughly constant, it is the intensity pattern that will change: light precipitation events decrease and heavy precipitation events increase. Droughts and heat waves are assumed to become more frequent, longer and more intense, also as a feedback mechanism of reduced soil moisture affecting evapotranspiration. In addition to meteorological droughts, edaphic droughts are anticipated to increase in the future. These developments impact the soil hydrological functions with altered infiltration conditions, increased surface runoff and an increasing proportion of preferential flow affecting a more complex and heterogeneous water distribution in the subsurface. Yet the link between tree mortality and the reduced and more heterogenous soil water distribution is still not fully understoodThe majority of approaches analysing soil moisture dynamics are based on point measurements, which do not account for the high spatial variability of soil water. Here, we close this knowledge gap by fusing established point-measurements with geophysical methods to assess the spatio-temporal dynamics of water fluxes in the near-surface subsoil from slope to the root zone scale. The questions we ask focus on how infiltration, subsurface water flow, soil moisture distribution and persistence are affected by (i) the subsurface architecture including textural variations as well as preferential flow paths (macro pores, root tracks) and (ii) hydrological extremes (droughts, rain events).Our study site is located in a beech forest near Ebergötzen (central Germany). The Triassic sandstones are overlain by periglacial slope deposits with varying amounts of loess. The Ebergötzen test site is equipped with numerous sensors for analysing water and element fluxes. In addition to meteorological parameters, we collect 15 min times series of throughfall, stemflow, soil water content, water tension and sap flow. This set-up is ideally suited to quantify water fluxes on a point-by-point basis with high temporal resolution, and to validate complementary, beyond-point approaches. To account for the small-scale variability of processes, geophysical methods with a focus on high-resolution electrical resistivity tomography (Dipole-Dipole, 48 electrodes, 15 cm spacing) were used. Measurements were carried out as a combination of a long-term approach (fortnightly/monthly) and event-based measurements (thunderstorm, round the clock).Our data indicate a relatively uniform decrease in soil moisture during prolonged dry periods, with root-water uptake locally causing higher dynamics. In contrast, subsurface moisture penetration after precipitation events is spatially highly variable, confirming the importance of preferential flow for infiltration and distribution of water in the subsurface and thus show the high demand for spatially high-resolution measurements of soil moisture dynamics.

The ability to monitor, understand, and predict the dynamics of the terrestrial carbon cycle requires the capacity to robustly and coherently synthesize multiple streams of information that each provide partial information about different pools and fluxes. In this study, we introduce a new terrestrial carbon cycle data assimilation system, built on the PEcAn model–data eco-informatics system, and its application for the development of a proof-of-concept carbon “reanalysis” product that harmonizes carbon pools (leaf, wood, soil) and fluxes (GPP, Ra, Rh, NEE) across the contiguous United States from 1986–2019. We first calibrated this system against plant trait and flux tower net ecosystem exchange (NEE) using a novel emulated hierarchical Bayesian approach. Next, we extended the Tobit–Wishart ensemble filter (TWEnF) state data assimilation (SDA) framework, a generalization of the common ensemble Kalman filter which accounts for censored data and provides a fully Bayesian estimate of model process error, to a regional-scale system with a calibrated localization. Combined with additional workflows for propagating parameter, initial condition, and driver uncertainty, this represents the most complete and robust uncertainty accounting available for terrestrial carbon models. Our initial reanalysis was run on an irregular grid of ∼ 500 points selected using a stratified sampling method to efficiently capture environmental heterogeneity. Remotely sensed observations of aboveground biomass (Landsat LandTrendr) and leaf area index (LAI) (MODIS MOD15) were sequentially assimilated into the SIPNET model. Reanalysis soil carbon, which was indirectly constrained based on modeled covariances, showed general agreement with SoilGrids, an independent soil carbon data product. Reanalysis NEE, which was constrained based on posterior ensemble weights, also showed good agreement with eddy flux tower NEE and reduced root mean square error (RMSE) compared to the calibrated forecast. Ultimately, PEcAn's new open-source regional data assimilation framework provides a scalable workflow for harmonizing multiple data constraints and providing a uniform synthetic platform for carbon monitoring, reporting, and verification (MRV) as well as accelerating terrestrial carbon cycle research.

Exceptional rainfalls (up to 200 mm in 72 h) in W-Germany, the Netherlands and Belgium led to severe flooding on 14-15 July 2021. In Germany the Ahr valley (Eifel mountains) was hit heavily, leading to 134 fatalities and substantial loss of property and infrastructure. Besides the damage in the floodplains, multiple shallow landslides were triggered along the Ahr embankments. Furthermore, the flood caused undercutting of several old landslide bodies. One such landslide in Devonian Schist bedrock is located at a narrow, bended stretch of the Ahr, near the town of Müsch. A complete failure has the potential to dam the river posing a considerable hazard.The main objectives of this study are to gain an in-depth understanding of the landslide causes and its transient activity. These objectives are tackled by a multi-method approach: landslide mapping, analysis of pre- and post-event airborne laser scanning (ALS) and terrestrial laser scanning (TLS) data, electrical resistivity tomography (ERT), seismic refraction tomography (SRT), passive seismic monitoring, geotechnical analysis and interviews with local inhabitants.The old landslide is 100 m wide and 200 m long. Preliminary analysis of ERT and SRT indicate a landslide depth of 20-30 m, leading to an overall landslide volume of 400,000 - 600,000 m³. ERT further shows underlying bedrock properties and water saturated zones. An old dumpsite as well as an ancient railway, now used as forest trail, cutting through the landslide horizontally are clearly shown as resistive zones. Analysis of ALS data shows that so far only the frontal part at the Ahr banks has been active and has lost about 6300 m³ due to fluvial erosion and landsliding. Field mapping shows clear signs of retrogressive landsliding. From October 2021 onwards the landslide body has been equipped with five geophones to record both subtle changes in ground rheology and discrete events of rock bridge failure due to incremental mass movement. Currently most seismic signals at the slope can be allocated to daily traffic and road construction in the area.The combination of geophysical and remote sensing methods enables a profound insight into the mechanisms and present processes of the Müsch landslide. Based on this, we will be able to assess the probability for a reactivation of the whole landslide body, which could trigger cascading hazards affecting a much larger region. An improved monitoring concept will be developed which can be adopted to similar structures in the Ahr valley and beyond.

Valley confined floods are a major hazard. In contrast to large river floods with day long warning time, they can evolve within minutes to hours, exhibit higher flow velocities and drive large amounts of debris into populated places. While many Alpine communities have developed mitigation, early warning and rapid response schemes for this natural hazard type, these measures are virtually unknown in Central European upland regions. Beyond flood protection, lacking measurement infrastructure also prevents retrospective collection of event anatomy data, which would be key to understand the evolution of an event and, hence to improve our response to future hazards.The 14–15 July 2021 flood that hit the Ahr valley in the Eifel mountains, west Germany, was a drastic example of the potential of such valley confined floods. A wall of water flushed through the deeply incised valley, flooding more than 15 towns and affecting 42,000 people, resulting in the highest number of casualties in Germany since 1962. All gauges along the main channel were destroyed while the flood hydrograph was still on the rising limb and grid power loss interrupted collection and transmission of data from other potential sensors.Here, we use data from a single seismic station near the town of Ahrweiler, originally deployed for earthquake seismology. Despite grid power cutoff around 23:19 CEST, the station recorded the arrival of the fast rising limb of the flood. We show how even an incomplete record of a single station not set up for flood early warning can be used to infer crucial and timely information about the flood: propagation velocity, water level and debris transport rate. We argue that installing a network of a few distributed low cost seismic sensors could have improved flood early warning and near real time provision of kinetic flood data. More importantly, such a network would be the key for improved response actions for future floods, deemed more likely in Central Europe under the currently changing climate conditions.

Bed load transport is a critical parameter in the study of landscape evolution and also provides valuable information for problems in the fields of ecology, river and landuse management, and civil engineering. Bed load transport is difficult to assess due to its stochastic nature and highly variable transport rates, and traditional measurement techniques have struggled to capture the spatial and temporal variability of bed load transport. In recent years, bed load monitoring based on seismological observations has emerged, which allows non-invasive and continuous indirect measurements. However, there still remains a significant challenge to independently characterise the seismic signature of bed load from other sources of noise, such as turbulence. Our study aims to explore seismic data recorded at the highly braided River Feshie in Scotland, which has undergone significant morphological change in its history and has been highly monitored over the last couple decades through Digital Elevation Models. Since the deployment of our seismometers in December 2020 we have captured three independent high flow events plus an isolated earthquake, which are being used to determine the environmental signals and the site specific signal characteristics. In some previous studies, an observed hysteretic relationship between seismic power and hydrological parameters has been interpreted as being characteristic of bed load transport. From the data we have gathered we have observed a hysteresis in the signals, and through Shields calculations it is suggested that bed load transport would be expected during these events. However, without independent constraints we do not feel we can be absolutely certain that this behaviour is a result of bed load transport. Our ongoing study therefore aims to combine multiple physical measurement techniques, such as hydroacoustic measurements, time-lapse imagery and seismic observations, to try and pinpoint what is contributing to the seismic signals recorded and how we can isolate the bed load transport component.

In order to reduce the societal impact of mass-wasting events, we need observations to investigate the factors that control slope failure, such as the state of crack propagation along a failure plane. However, usually the failure plane is not accessible in-situ. Hence, cracks have to be monitored indirectly, for example using seismic methods.We analysed the data from a seismometer array in the Illgraben catchment, Switzerland, that had registered a series of crack propagation and mass-wasting events, leading to a main event that happened on 2 January 2013. We used a state-of-the-art machine learning technique based on hidden Markov models to detect and classify the seismic signals of crack events. We obtained the temporal evolution of three signal types: (1) single crack signal, (2) rock avalanche and (3) rockfall activity due to debris remobilization. The temporal evolution of the number of cracks showed a linear trend in the weeks prior to the main mass-wasting event and, in the hours preceding the main event, a sigmoidal exponential growth. Using these observations, we propose a mechanistic model to describe the rupture of the failure plane. The model considers the internal parameter of the total crack boundary length as the primary control on failure plane evolution, in addition to the previously suggested crack propagation velocity control parameter. According to this model, internal parameters appear to be the dominant control for the failure plane growth at a slope scale.

The ability to monitor, understand, and predict the dynamics of the terrestrial carbon cycle requires the capacity to robustly and coherently synthesize multiple streams of information that each provide partial information about different pools and fluxes. In this study, we introduce a new terrestrial carbon cycle data assimilation system, built on the PEcAn model-data eco-informatics system, and its application for the development of a proof-of-concept carbon "reanalysis" product that harmonizes carbon pools (leaf, wood, soil) and fluxes (GPP, Ra, Rh, NEE) across the contiguous United States from 1986–2019. We first calibrated this system against plant trait and flux tower Net Ecosystem Exchange (NEE) using a novel emulated hierarchical Bayesian approach. Next, we extended the Tobit-Wishart Ensemble Filter (TWEnF) State Data Assimilation (SDA) framework, a generalization of the common Ensemble Kalman Filter which accounts for censored data and provides a fully Bayesian estimate of model process error, to a regional-scale system with a calibrated localization. Combined with additional workflows for propagating parameter, initial condition, and driver uncertainty, this represents the most complete and robust uncertainty accounting available for terrestrial carbon models. Our initial reanalysis was run on an irregular grid of ~500 points selected using a stratified sampling method to efficiently capture environmental heterogeneity. Remotely sensed observations of aboveground biomass (Landsat LandTrendr) and LAI (MODIS MOD15) were sequentially assimilated into the SIPNET model. Reanalysis soil carbon, which was indirectly constrained based on modeled covariances, showed general agreement with SoilGrids, an independent soil carbon data product. Reanalysis NEE, which was constrained based on posterior ensemble weights, also showed good agreement with eddy flux tower NEE and reduced RMSE compared to the calibrated forecast. Ultimately, PEcAn's carbon cycle reanalysis provides a scalable framework for harmonizing multiple data constraints and providing a uniform synthetic platform for carbon monitoring, reporting, and verification (MRV) and accelerating terrestrial carbon cycle research.

The Ecosystem Demography model version 2.2 (ED-2.2) is a terrestrial biosphere model that simulates the biophysical, ecological, and biogeochemical dynamics of vertically and horizontally heterogeneous terrestrial ecosystems. In a companion paper (Longo et al., 2019a), we described how the model solves the energy, water, and carbon cycles, and verified the high degree of conservation of these properties in long-term simulations that include long-term (multi-decadal) vegetation dynamics. Here, we present a detailed assessment of the model's ability to represent multiple processes associated with the biophysical and biogeochemical cycles in Amazon forests. We use multiple measurements from eddy covariance towers, forest inventory plots, and regional remote-sensing products to assess the model's ability to represent biophysical, physiological, and ecological processes at multiple timescales, ranging from subdaily to century long. The ED-2.2 model accurately describes the vertical distribution of light, water fluxes, and the storage of water, energy, and carbon in the canopy air space, the regional distribution of biomass in tropical South America, and the variability of biomass as a function of environmental drivers. In addition, ED-2.2 qualitatively captures several emergent properties of the ecosystem found in observations, specifically observed relationships between aboveground biomass, mortality rates, and wood density; however, the slopes of these relationships were not accurately captured. We also identified several limitations, including the model's tendency to overestimate the magnitude and seasonality of heterotrophic respiration and to overestimate growth rates in a nutrient-poor tropical site. The evaluation presented here highlights the potential of incorporating structural and functional heterogeneity within biomes in Earth system models (ESMs) and to realistically represent their impacts on energy, water, and carbon cycles. We also identify several priorities for further model development.

Earth system models (ESMs) have been developed to represent the role of terrestrial ecosystems on the energy, water, and carbon cycles. However, many ESMs still lack representation of within-ecosystem heterogeneity and diversity. In this paper, we present the Ecosystem Demography model version 2.2 (ED-2.2). In ED-2.2, the biophysical and physiological processes account for the horizontal and vertical heterogeneity of the ecosystem: the energy, water, and carbon cycles are solved separately for a series of vegetation cohorts (groups of individual plants of similar size and plant functional type) distributed across a series of spatially implicit patches (representing collections of micro-environments that have a similar disturbance history). We define the equations that describe the energy, water, and carbon cycles in terms of total energy, water, and carbon, which simplifies the differential equations and guarantees excellent conservation of these quantities in long-term simulation (< 0.1 % error over 50 years). We also show examples of ED-2.2 simulation results at single sites and across tropical South America. These results demonstrate the model's ability to characterize the variability of ecosystem structure, composition, and functioning both at stand and continental scales. A detailed model evaluation was conducted and is presented in a companion paper (Longo et al., 2019a). Finally, we highlight some of the ongoing model developments designed to improve the model's accuracy and performance and to include processes hitherto not represented in the model.

The slope instability and associated mass wasting are among the most efficient surface gradation processes in the bedrock terrain that produce dramatic landscape change and associated hazards. The wedge failure in periglacial Higher Himalaya terrain on 7th February in Chamoli, Uttarakhand (India) produced >1.5 km high rock avalanche, which amalgamated with the glacial debris on the frozen river bed produced massive debris flow along the high gradient Rishi Ganga catchment. The high-velocity debris flow and a surge of high flood led to extensive loss of life and infrastructures and issuing the extreme event flood warning along the Alakananda-Ganga river, despite there was no immediate extreme climatic event. The affected region is the locus of extreme mass wasting events associated with Glacial Lake Outburst Flood (GLOF) and Landslide Lake Outburst Flood (LLOF) in the recent past. We analyzed the landscape to understand its control on the 7th February 2021 Rishi Ganga event and briefly discuss other significant events in the adjoining region e.g. 1893/1970 Gohna Tal/Lake LLOF and 2013-Uttarakhand events in Chamoli, which have significance in understanding the surface processes in Higher Himalayan terrain.

Large rock slope failures play a pivotal role in long-term landscape evolution and are a major concern in land use planning and hazard aspects. While the failure phase and the time immediately prior to failure are increasingly well studied, the nature of the preparation phase remains enigmatic. This knowledge gap is to a large degree related to challenges in collecting appropriate data in such high mountain terrain. Classic monitoring techniques provide detailed data but mostly of point character and only reflecting the surface expression of processes within the rock mass. Thus, the integral behaviour of a peak, at the surface and at depth remains elusive.Here, we present results from a continuous multi-sensor seismic analysis of the Hochvogel summit, a 2592 m high Alpine peak, which is deemed to fail in the near future, as a 5 m wide and 40 m long crack is progressively opening and mobilising up to 260,000 cubic metres of rock. The seismic network consisted of up to seven sensors, installed during July--October 2018 (with 43 days of data loss). We develop and discuss proxy time series indicative of cyclic and progressive changes of the summit.Modal analysis, horizontal-to-vertical spectral ratio data and end-member modelling analysis reveal diurnal cycles of increasing and decreasing coupling stiffness of the fragmented rock volume, due to thermal forcing. Relative seismic wave velocity changes mimic this pattern but also reveal the release of stress within the rock mass. At longer time scales, there is a superimposed pattern of stress evolution, which increases for five to seven days and suddenly drops within a few days, also expressed in an increased emission of short seismic pulses indicative of rock cracking. Our data provide essential first order information on an early stage of a large-scale slope instability, which evolves towards a catastrophic failure.

Environmental seismology is the discipline that uses ambient noise to detect and to measure geomorphic processes. The basic principle relies on the unique seismic signal, in terms of excited frequencies and amplitudes, generated by such processes which is then propagated and recorded to sensors (geophones). Recent developments of this technique are interesting for geomorphology because there is evidence that it can be used to study processes that are rare and difficult to measure, such as bedload transport in rivers. This is of particular importance for quantify bedload flux in proglacial streams. This paper focuses on using this method to quantify bedload export from an Alpine valley glacier. At present, there is no time-continuous measurement of bedload export in such settings. This is because objective measurement of bedload flux is a challenge task for both theoretical and practical reasons. Theoretically, the high variability of bedload transport means that measurements need to be continuous to identify when it happens and with which intensity. Practically, in order to know how to sample the signal through time, you need to know its variance a priori, which is commonly unsteady, and therefore needs to be sampled in order to know how to sample it. Direct manual sampling (e.g. with Helley-Smith samplers) have a serious disadvantage in this sense. Existing indirect methods (e.g. hydrophones and plate geophones) allow collection of continuous data but, at the same time, their installation can be difficult and expensive and they need to be carefully calibrated against manual samplings.In the present study we used seismic data collected by out-of-stream geophones to infer bedload flux along an entire melt season (June - September) at the Glacier d’Otemma proglacial forefield, south-western Swiss Alps. Raw seismic data were inverted into sediment flux following the methodology developed in Dietze et al. (2019), to produce the first continuous dataset of bedload export from an Alpine glacier. The inversion model has been calibrated using statistical (i.e. sensitivity analysis) and direct (i.e. in situ active seismic survey) approaches.

Desert research has a bold track record in deciphering the evolution of landforms. While iconic objects such as inselbergs, yardangs, wadis or dune fields have attracted the attention of scientists for decades and inspired them to develop numerous formation concepts, it is the actual processes, which are suggested to govern their formation, that are usually quite hard to survey as they happen in nature. This amiguitity is due to the often unpredictable, event based, rapid yet episodic nature of their occurrence. While dedicated devices for measuring these processes inevitably exist, they are either confronted with a small spatial footprint (point measurement) or temporal discontinuousity (time lapse data), or even both of these drawbacks. As a consequence, important processes like flash floods, bedload agitation, mass movement on hillslopes, thermal weathering, dune migration rate and mechanism, or averaged information about ground moisture change, substrate rheology, and temperature diffusion can be studied as they happen under natural conditions only in exceptional cases.Here I summarise recent developments from the field of environmental seismology, an emerging discipline that studies the seismic signals emitted by Earth surface processes, with a focus on existing and potential applications to desert environments. I present seismic approaches to detecting, tracking and quantifying flash floods and the sediment they convey, provide an overview of the capabilites of seismology to survey hillslope activity at the catchment scale, and discuss the potential to monitor surface and subsurface activity at the plot scale. The overarching aim of the contribution is to foster a discussion about potential research questions that could be tackled in future joint (i.e., established and seismic sensor) instrumentation projects.

Rock glaciers are typically regarded as periglacial features and their dynamics are supposed to be driven by ice content. Under ongoing global warming we expect these structures and dynamics to change and at least decay. This would be especially the case of rock glaciers in climate-sensitive high mountains of the Tibetan plateau, like in the Nyainqêntanglha range. Despite the similar past and present periglacial climatic conditions in this region, rock glaciers are only formed in a few, specific valleys. With this study, we aim to provide insights into the environmental conditions under which rock glaciers are formed and maintained, to be able to better understand how they will respond to changing boundary conditions, imposed by global warming.To assess “what makes a rock glacier?” we studied such a feature in the Qugaqie basin, at 5500 m a.s.l. To describe the structure and the dynamics of this active rock glacier we applied several methods (geomorphological mapping, geophysics, remote sensing) and we incorporated catchment area properties such as geology, water and sediment sources. Mapping of the geomorphology, the geology and surface material properties characterizes the external structure of the rock glacier. The internal structure, like the active layer zone and the existence of ice, is described by electrical resistivity tomography (ERT). To investigate the surface dynamics of the rock glaciers, we quantify displacement rates using Interferometric Synthetic Aperture Radar (InSAR) time-series analysis. To gain insight to internal deformation dynamics we use environmental seismology, allowing for detection and location of crack signals within the rock glacier. The seismic network also allows tracking rock falls at the head scarp and continuously monitoring glaciofluvial patterns. We find that the singularity of the presence of the studied rock glacier is most likely related to a specific melange of the geological structures, former glaciation of the valley, catchment size and shape and especially water availability.

Despite their low contribution to forest carbon stocks, lianas (woody vines) play an important role in the carbon dynamics of tropical forests where they compete with free-standing plants for below- and above-ground resources. Doing so, they negatively impact individual tree growth, as well as the net productivity and the long-term carbon storage of the ecosystem.However, lianas remain largely ignored in field-scale studies as well as modelling forecasts. Therefore, their exact impact on tropical forest biogeochemical cycles is very uncertain. In particular, it is unclear which resource (light, water) is the most competed for between growth forms and so is is the future impact of lianas on forests in a global climate change context in which brighter, drier and CO2-enriched conditions are expected in the Tropics.To answer those burning questions, we incorporated for the very first time a plant functional type accounting for the lianescent growth form into a dynamic global vegetation model (ED2). We implemented several liana-specific processes in the modelling framework (climbing, resprouting, height limitation due to lack of self-supporting tissues etc.), and integrated liana-specific parameters according to data from multiple studies in order to account for significant differences of functional and structural traits between lianas and trees. These parameters included (but were not limited to) leaf biochemical and photosynthesis properties, stem hydraulic traits, root distribution, and allometric relationships.Baseline runs successfully reproduced ecosystem gas exchange fluxes (GPP and latent heat), forest structural features (LAI, AGB), and several other benchmarking observations in multiple tropical sites characterized by different rainfall regimes and levels of liana abundance. In those simulations, lianas negatively reduced forest productivity and total carbon storage, by increasing tree mortality (+ 30% on average) and decreasing tree growth (-35%). The inclusion of lianas in the simulations reduced the forest net productivity by up to 0.5 tC ha−1 year−1, which resulted in significantly reduced accumulated above‐ground biomass by up to 20 tC/ha in regrowth forests. The negative impact of lianas on carbon storage almost disappeared in wetter, old-growth forest sites. Model uncertainty analyses also revealed that water limitation was the dominant factor driving competition between trees and lianas, even in sites with a short dry season.These two-key findings (higher impact in regrowth forests and water-dominated competition) are expected to lead to a reinforcement of the negative impact of lianas on forest productivity under future aggravated forest disturbance and warmer climate conditions. The modelling workflow also allowed to identify key liana traits (quantum efficiency, stomatal regulation parameters, allometric relationships) and processes (water use, respiration, climbing) driving the overall model uncertainty. They should be considered as priorities for future data acquisition and model development to improve predictions of liana-infested forest carbon dynamics.

High resolution topographic models generated from repeat unmanned aerial vehicle (UAV) surveys and structure from motion (SfM) are increasingly being used to investigate landscape changes and geomorphic processes. Traditionally, accurate UAV surveys require the use of independently measured ground control points or highly accurate camera position measurements. However, in addition to accuracy in an absolute sense (how well modeled topography matches real topography), model quality can be expressed as accuracy in a comparative sense (the degree to which two models match each other). We present a simple SfM workflow for calculating pairs or sets of models with a high comparative accuracy, without the need for ground control points or a dGNSS equipped UAV. The method is based on the automated detection of common tie points in stable portions of the survey area and, compared to a standard SfM approach without ground control, reduces the level of change detection in our surveys from several meters to 10-15 cm. We apply this approach in a multi-year monitoring campaign of an 8 km stretch of coastal cliffs on the island of Rügen, Germany. We are able to detect numerous mass wasting events as well as bands of more diffuse erosion in chalk sections of the cliff. Both the cliff collapses and the diffuse erosion appear to be strongly influenced by antecedent precipitation over seasonal timescales, with much greater activity during the winter of 2017-2018, following an above average wet summer, than during the subsequent two winters, which both followed relatively dry summers. This points to the influence of subsurface water storage in modulating cliff erosion on Rügen.

Forest albedo changes with vegetation dynamics, ecosystem demography and succession as different plant species can be characterized by contrasting leaf traits, as well as different allocation strategies. In tropical ecosystems, lianas (woody vines) strongly impact the forest biogeochemical cycles by competing with co-occurring trees for above- and below-ground resources. In addition to the particular location of their foliage (often at the top of the canopy), lianas were shown to contrast with tropical trees in terms of biochemical and structural leaf properties, as well as allocation strategies. As a consequence, liana spectral signature differs in several regions of the leaf spectrum (visible, near infrared and shortwave infrared) from tree’s one. At larger scale, the forest canopy reflectance spectrum is also affected by the relative abundance of lianas.To evaluate the impact of lianas on the radiative transfer and albedo of tropical forests, we collected all published reflectance spectra of liana and co-occurring tree leaves as well as the canopy reflectance spectra characterized by high and low liana coverage. We then calibrated both a leaf (PROSPECT-5) and a canopy (ED2-RTM) radiative transfer model on those data to reproduce the spectral signatures together with their differences, for each single study and site. The Bayesian framework that we used generated leaf biochemical and structural trait distributions, as well as allocation pattern strategies that could be compared between both growth forms.Collected spectra could be fairly well reproduced at both leaf and canopy levels by the selected mechanistic models. In most studies, calibration led to significantly lower chlorophyll and carotenoid contents, higher relative water content, and larger specific leaf areas for liana leaves, which allowed reproducing observed higher leaf reflectance values in the visible and shortwave infrared and the lower ones in the near infrared. We then validated our findings with independent field data on leaf chemistry and structure from the same studies.These calibrated radiative transfer models are tools that can be used in future research to investigate the liana impact on the global energy budget of tropical forests, as well as to monitor spatial and temporal liana abundance.

Constraining bedload flux in rivers is a challenging objective, especially when the data need to be continuous, beyond-point estimates. Seismometers are potentially valuable alternatives to in‐stream devices, which involve extensive measurement infrastructure or labour‐intensive manual sampling that can be potentially dangerous. We present a Monte Carlo-based inverse approach to deducing hydraulic and bedload transport dynamics continuously, with high temporal resolution, from seismic data, that averages the system’s behaviour over tens of metres. Water depths and bedload fluxes can be reproduced with average deviations of 0.10 m and 0.02 kg/sm, respectively. The method is validated against synthetic data sets and independently measured metrics from several challenging streams: we show applications of the technique from a flash flood-dominated catchment in Israel (Nahal Eshtemoa), from an ice-covered subarctic river (Sävarån, Sweden), and from a typhoon-driven major mountain river in Taiwan (Liwu River). The presented approach is a generic method implemented in the R package ‘eseis’ that can be used with off-the-shelf seismic equipment, installed at safe distances from potentially hostile conditions with minimum site disturbance.

Process-based vegetation models are widely used to predict local and global ecosystem dynamics and climate change impacts. Due to their complexity, they require careful parameterization and evaluation to ensure that projections are accurate and reliable. The PROFOUND Database (PROFOUND DB) provides a wide range of empirical data to calibrate and evaluate vegetation models that simulate climate impacts at the forest stand scale. A particular advantage of this database is its wide coverage of multiple data sources at different hierarchical and temporal scales, together with environmental driving data as well as the latest climate scenarios. Specifically, the PROFOUND DB provides general site descriptions, soil, climate, CO2, nitrogen deposition, tree and forest stand-level, as well as remote sensing data for nine contrasting forest stands distributed across Europe. Moreover, for a subset of five sites, time series of carbon fluxes, atmospheric heat conduction, and soil water are also available. The climate and nitrogen deposition data contain several datasets for the historic period and a wide range of future climate change scenarios following the Representative Concentration Pathways (RCP2.6, RCP4.5, RCP6.0, RCP8.5). We also provide pre-industrial climate simulations that allow for model runs aimed at disentangling the contribution of climate change to observed forest productivity changes. The PROFOUND DB is available freely as a SQLite relational database or ASCII flat file version (at https://doi.org/10.5880/PIK.2019.008). The data policies of the individual, contributing datasets are provided in the metadata of each data file. The PROFOUND DB can also be accessed via the ProfoundData R-package (https://github.com/COST-FP1304-PROFOUND/ProfoundData), which provides basic functions to explore, plot, and extract the data for model set-up, calibration and evaluation.

Environmental seismology is the study of the seismic signals emitted by Earth surface processes. This emerging research field is at the seams of seismology, geomorphology, hydrology, meteorology, and further Earth science disciplines. It amalgamates a wide variety of methods from across these disciplines and, ultimately, fuses them in a common analysis environment. This overarching scope of environmental seismology asks for a coherent, yet integrative software, which is accepted by many of the involved scientific disciplines. The statistic software R has gained paramount importance in the majority of data science research fields. R has well justified advances over other, mostly commercial software, which makes it the ideal language to base a comprehensive analysis toolbox on. The article introduces the avenues and needs of environmental seismology, and how these are met by the R package eseis. The conceptual structure, example data sets and available functions are demonstrated. Worked examples illustrate possible applications of the package and in depth descriptions of the flexible use of the functions. The package is available under the GPL license on the Comprehensive R Archive Network (CRAN) and maintained on Github.

Rockfalls are an essential geomorphic process and an important natural hazard in steep landscapes across the globe. Seismic monitoring can provide precise information on the timing, location and event anatomy of rockfalls, parameters that are otherwise hard to constrain. By pairing data from 49 seismically detected rockfalls in the Lauterbrunnen Valley, Swiss Alps, with independent information about potential triggers during autumn 2014 and spring 2015, we are able to (i) analyse the evolution of single rockfalls and their common properties, (ii) identify seasonally changing activity hotspots, (iii) and explore temporal activity patterns at different scales, ranging from months to minutes, to quantify relevant trigger mechanisms. Seismic data allows the classification of rockfall activity into three distinct phenomenological types and can be used to discern multiple rock mass releases from the same spot, identify rockfalls that trigger further rockfalls and resolve modes of subsequent talus slope activity. In contrast to findings based on methods with longer integration times, rockfall in the monitored limestone cliff is not spatially uniform but shows a downward shift of rock mass release spots by 33 m per month over the year, most likely driven by a continuously lowering water table. Freeze-thaw-transitions account for only 5 out of the 49 rockfalls whereas 19 rockfalls were triggered by rainfall events, with a peak lag time of 1 h. Another 17 rockfalls were triggered by diurnal temperature changes and occurred during the coldest hours of the day as well as during the highest temperature change rates. This study is thus the first one to show direct links between proposed rockfall triggers and the spatio-temporal distribution of rockfalls under natural conditions, and extends existing models by providing seismic observations of the rockfall process prior to the first rock mass impacts.

Rockfall in deglaciated mountain valleys is perhaps the most important post-glacial geomorphic process for determining the rates and patterns of valley wall erosion. Furthermore, rockfall poses a significant hazard to inhabitants and motivates the monitoring for rockfall occurrence in populated areas. Traditional rockfall detection methods, such as aerial photography and Terrestrial Laser Scanning (TLS) data evaluation provide constraints on the location and released volume of rock, but have limitations due to significant time lags or integration times between surveys, and deliver limited information on rockfall triggering mechanisms and the dynamics of individual events. Environmental seismology, the study of seismic signals emitted by processes at the Earth's surface, provides a complementary solution to these shortcomings. This approach is limited amongst others by the strength of the signals emitted by a source and their transformation and attenuation towards receivers. To test the ability of seismic methods to identify and locate small rockfalls, and to characterise their dynamics, we surveyed a 2.16 km2 large, near vertical cliff section of the Lauterbrunnen Valley in the Swiss Alps with a TLS and six broadband seismometers. During 37 days in autumn 2014, ten TLS-detected rockfalls with volumes ranging from 0.053 &pm; 0.004 to 2.338 &pm; 0.085 m3 were independently detected and located by the seismic approach, with a deviation of 81−29+59 m (about 7 % of the average inter-station distance of the seismometer network). Further potential rockfalls were detected outside the TLS-surveyed cliff area. The onset of individual events can be determined within a few milliseconds, and their dynamics can be resolved into distinct phases, such as detachment, free fall, intermittent impact, fragmentation, arrival at the talus slope and subsequent slope activity. The small rockfall volumes in this area require significant supervision during data processing: 2175 initially picked potential events reduced to 511 potential events after applying automatic rejection criteria. The 511 events needed to be inspected manually to reveal 19 short earthquakes and 37 potential rockfalls, including the ten TLS-detected events. Rockfalls do not show a relationship to released seismic energy or peak amplitude at this spatial scale due to the dominance of process-inherent factors, such as fall height, degree of fragmentation and distribution, and subsequent talus slope activity. The combination of TLS and environmental seismology provides a detailed validation of seismic detection of small volume rockfalls, and revealed unprecedented temporal, spatial and geometric details about rockfalls in steep mountainous terrain.