Coastal cities face recurrent flooding from storm events and rising seas. A contributing factor to flooding in these low relief areas is the groundwater table, which, already relatively shallow, can quickly rise towards the land surface during storm events. This leads to increased surface runoff entering stormwater drainage systems and a greater probability of flooding. As such, groundwater table forecasts could be an important component of real-time flood forecasting systems, but are generally unavailable. Because traditional physics-based models require extensive amounts of subsurface data that is difficult to obtain, especially in urban environments, this research evaluates two types of machine learning models, Recurrent Neural Networks (RNN) and Long Short-term Memory neural networks (LSTM), for creating groundwater table forecasts. The two types of networks were built with Tensorflow/Keras to forecast the groundwater table response to forecasted storm events and appropriate hyperparameters were tuned using the Hyperas library. Using observed hourly groundwater levels, rainfall, and tide from the City of Norfolk, Virginia, the networks were trained with data from 2010-2016 and tested with data from 2016-2018. Archived forecast rainfall and tide from two large storms in the test period (Hurricane Hermine and Tropical Storm Julia) were then used to evaluate the effect of forecast inputs on model performance. Results indicate that LSTM is slightly more accurate when forecasting the groundwater table than RNN, likely because of its increased ability to preserve and learn from past information. Average root mean squared error and Nash-Sutcliffe efficiency values for an 18hr forecast for the LSTM were 0.06m and 0.89, respectively, and 0.07m and 0.85, respectively, for the RNN. These forecasts could provide valuable information to aid in planning and response to storm events and will become an increasingly important part of effectively modeling and predicting coastal urban flooding as sea level rises.

Regional Climate Models (RCMs) work at finer resolution over a limited region and are presumed to perform better at regional scales. RCMs need thorough evaluation before being used for any climate change impact assessment study due to the biases associated with the observed data. While few studies used RCM outputs for understanding the spatio-temporal variability of precipitation and temperature over India, application of RCMs in drought assessment has been overlooked. Here, the study aims to perform drought analysis using RCMs over India with Standardized Precipitation Evapotranspiration Index (SPEI) as the drought index. About 10 RCMs from the Coordinated Regional Climate Downscaling Experiment program (CORDEX) have been considered in the analysis. To remove the systematic biases, a Quantile based bias correction method has been used. The study evaluated the performance of bias-corrected RCMs to simulate rainfall over India for each grid using the statistical measures such as correlation and Nash-Sutcliffe Efficiency coefficients. The monthly precipitation for all over India was best represented by the experiment LMDz-IITMRegCM4 (Regional Climatic Model version 4). Based on the performance evaluation in the study, ICHEC-EC-EARTH-SMHI-RCA4 and MPI-CSC-REMO2009 were used along with LMDz-IITMRegCM4 for drought assessment over India. The results reveal that for West and North-East zones, the drought frequencies and intensities increase for the periods of 2001-2050 and 2051-2100 with Representative Concentration Pathways (RCP) 4.5 for all three considered RCMs. All over India, the average drought intensities were observed to be increasing for ICHEC-EC-EARTH-SMHI-RCA4 and LMDz-IITMRegCM4 while there is no change for MPI-CSC-REMO2009.

Availability of far ultraviolet observations of Earth’s dayglow by the NASA Global-scale Observations of the Limb and Disk (GOLD) mission presents an unparalleled opportunity for upper atmosphere data assimilation. Assimilation of the observed dayglow emissions can be formulated in a similar fashion to lower atmosphere radiance data assimilation approaches using the sensitivity of the Lyman-Birge-Hopfield (LBH) band emission to thermospheric temperature. To demonstrate such an approach, we present a proof-of-concept implementation of an ensemble square-root filter measurement update step using ensemble simulation of the thermosphere and LBH emission by the NOAA’s Whole Atmosphere Model (WAM) and NCAR’s Global Airglow model. With help of a new assimilation approach, the utility of GOLD observations can be extended to reveal the global, time-dependent, altitude-resolved thermospheric structure, offering the key to addressing a number of outstanding questions such as origins of traveling atmospheric disturbances.

In the northeastern Colombia, the northernmost foothills of the Andes are present, whose representatives are the Sierra Nevada de Santa Marta (SNSM) and the Perija Range. This orogenic system is delimited by three major faults that limit three large basins. In its order from west to east are the Santa Marta Fault that limits the Sierra Nevada de Santa Marta with the Lower Magdalena Basin. The Oca Fault that limits with the low basin of the Rancheria River to the north in the south of Guajira Peninsula, and toward the east the Perija – El Tigre Fault that limits with the Maracaibo Basin. Each of these faults have a great size. Since 2008, when the National Seismological Network of Colombia (RSNC) increased its number of seismological stations in this region of Colombia, the recording of surface seismicity that is associated with the tectonic mobility of this orogenic system began. The strong earthquakes in this region do not exceed in magnitude M = 5.5, emphasizing earthquakes with M ≈ 4.5 in average. The origin of the tectonic mobility of this orogenic system obeys to the convergence between the Caribbean Plate and the northwestern corner of South America. As a historical antecedent in this region is the earthquake of May 22, 1834 that destroyed the city of Santa Marta. Along the western sector of the Oca fault, this is composed of parallel faults and towards the east when it crosses the basin of the Ranchería River, present evidences of neotectonics. This fault becomes a good candidate to produce an earthquake with magnitudes greater than 5.0. For the Santa Marta fault, the alteration of the landscape by anthropic effect has erased evidence of active tectonics in its northern sector, while towards SE it shows morphotectonics related to its fault plane, and the Perijá - El Tigre fault due to inaccessible conditions Because it is a jungle area, very large and lacking communication routes as well as having a long history of armed conflict, it has been impossible to study it in the field and it has only been verified from a morphotectonic point of view with the help of remote sensors. To this fault is attributed to the earthquake of September 11, 2014 with M = 4.1 and inverse focal mechanism according to the report of the RSNC. In this work a tectonic model is shown, the kinematic behavior of each fault, and its probable seismic risk for this region of Colombia.

River discharges that are large enough to be impacted by Coriolis when they reach the coastal ocean create a buoyancy-driven coastal current in the direction of Kelvin wave propagation. This geostrophically adjusted current is a major circulation feature, controlling the transport of riverine waters and associated materials, such as sediments, nutrients and pollutants. Coastal models in areas of large rivers need to accurately represent this coastal current and its variability under the influence of other factors impacting coastal circulation. Coastal altimetry data are an important source of suitable observations to guide modeling and applications. Use of such data around the Mississippi River Delta will be shown, in tandem with in situ data and a high resolution hydrodynamic model that has been optimized for river plume dynamics. During the 2010 Deepwater Horizon incident, this methodology evaluated a period of strong coastal current, which was found relevant to the transport of hydrocarbons from the historically unprecedented Gulf of Mexico oil spill. During the 2015 Mississippi flood episode, the coastal altimetry data confirmed the weak coastal current due to the offshore removal of riverine waters under the influence of oceanic circulation, namely the Loop Current and associated eddies. This variability was captured by both coastal altimetry data and the model, leading to good agreement with in situ data further offshore, along the branches that carried Mississippi waters toward South Florida.

We seek to calibrate the flow law for polythermal ice through shear strain analysis. In a warming climate, increased melting of glaciers and ice caps play a big role in sea level rise. Approximately 60% of the current contribution to sea level rise from ice loss is attributed to glaciers and ice caps, raising the urgency of sharpening mass balance change predictions in regions of streaming flow. Polythermal glaciers constitute a significant portion of these contributing glaciers, though our knowledge of their flow dynamics is incomplete. Thermally complex polythermal glaciers have both warm and cold ice which lead to weak wet-based beds, with significant amounts of basal sliding and deformable till. Consequently, polythermal glaciers experience significant shear strain as their lateral shear margins sustain the majority of the resisting stress. Most in-situ and in-lab studies of natural ice over recent years have focused on bodies of ice with frozen beds that experience minimal shear strain downglacier and across vertical planes (with depth) relative to the bed. The lack of studies on wet-based polythermal glaciers causes uncertainties in the flow law, as differences in flow law factors between polythermal ice and bodies of ice with frozen beds have the potential to induce more than an order of magnitude difference in ice velocity. To improve calibration of the flow law for polythermal ice, we seek to improve our understanding of their shear strain regimes. We developed and deployed tilt sensor systems on the polythermal Jarvis Glacier in Alaska, where we drilled multiple boreholes close to Jarvis’ shear margin and installed three boreholes with our tilt sensor systems. The tilt sensors measure gravity, magnetic and temperature data, and each system consists of multiple sensors connected along a cable and serially communicating along a common data bus with a datalogger. We have recently retrieved a year of Jarvis tilt sensor data and calculated the at-depth shear strain rates in the boreholes, allowing evaluation of the at-depth shear strain rate regimes of polythermal ice against theoretical models developed using Glen’s flow law. We present the development of our data collection methodology and the results of our shear strain analysis, with suggestions for potential calibrations of the flow law for polythermal ice.

In arid snow-fed river systems, where climate change affects snowpack accumulation and snowmelt timing, community-based participatory research that engages scientists and stakeholders can shape research agendas to support local climate adaptation and produce decision-relevant science. This presentation provides an overview of a community-based participatory research project underway in the Truckee-Carson River System in the Western United States. The authors: (1) describe the participatory research framework developed for this case study which features an interdisciplinary science team; (2) explain the selection and role of a diverse group of stakeholders representing diverse water use communities that regularly interacts with scientists; (3) highlight selected results of a local climate resiliency assessment and research activities to date; and (4) share successes and challenges to date useful to establishing best practices that may guide the replication of this research framework elsewhere. Our findings indicate that climate change is mobilizing local adaptation strategies that include communicating with other water managers, collecting data to monitor climate impacts, and planning for future water supply variability by investigating the performance of institutionalized water management regimes. To guide community decisions and support managers’ science information needs, researchers are developing stakeholder-informed climate scenarios and simulating locally identified adaptation strategies. Best practices identified to date underscore the importance of early stakeholder engagement to clarify case study boundaries, prioritize research questions of interest, and identify a core stakeholder group willing to participate in research. This sets the stage for a transparent and responsive process, which aids in building trust in the research design as demonstrated through scientists seeking and incorporating local knowledge and input. Related to this transparency is the need to acknowledge power disparities that may exist among stakeholder communities including historically marginalized groups with high stakes in sustaining water resources. A thorough stakeholder analysis should address these considerations and comprises a critical component of the research design.

The study and investigation of temperature trends is quite essential for a country like India, whose economy is largely dependent on agriculture as temperature greatly influences rainfall and evapotranspiration processes. Temperature along with other meteorological parameters largely affect the water availability over a region. Almost 83% water is used in agriculture, 12% is consumed by industries and merely 5% is used for domestic purposes. In the present study, long term trend with temperature data of 64 years (1951-2014) was analyzed for maximum, minimum and mean temperature values. After removal of all significant correlation in the data, Mann-Kendall test (MK) was applied. To identify the trends Sen’s slope estimator was used and the overall percent change for the temperature over the study area was determined. To interpolate the spatial pattern an inverse distance weighted interpolation method (IDW) was used using ArcGIS 10.2.2. The analysis showed a significant increase in the trend of temperature for annual and seasonal series during the entire period over the study area i.e. Kharun watershed, Chhattisgarh, India. These results will be quite helpful and play an important role in managing water resources in the basin and will be useful for future planning of irrigation and agricultural practices.

We present the first Northern Hemisphere high-resolution Holocene methane (CH4) record obtained from the Renland Ice Cap (ReCAP) – 71.3N 26.6W 2300 m a.s.l. in 2015. The total length of the core is 584 m containing 532 m of the Holocene ice without brittle ice zone thus allowing to obtain an uninterrupted CH4 record over the Holocene. An ice core analytical technique developed in 2010 – continuous flow analysis (CFA) provides a unique opportunity of revealing high-resolution greenhouse gas record of the past. Considering a mean annual layer thickness the depth resolution translates to a nominal temporal resolution of 35 data points 100 yr-1 over the Meghalayan (back to 4.2 ka BP) Holocene section. Note that due to the average residence time of CH4 in the atmosphere and the gas age distribution width in the firn, the maximum temporal effective resolution is lower by a factor of 10. The pattern of the centennial scale variability seen in the ReCAP CH4 record is coherent with the earlier published Southern Hemisphere record based on the West Antarctic Ice Divide (WAIS) CH4 record. The wavelet coherence analysis identifies a high correlation on the long- and midterm variability (600-700, 200-500 yr periodicity) as expected; the Meghalayan Holocene section reveals a common variability down to 70 yr. Inter-laboratory offsets of the absolute CH4 values are likely constant as the entire cores were running over continuing measurement campaigns. The elevated ReCAP CH4 level due to the local dust presence is disproved for at least the Last Glacial section. Gas trapping uncertainties should not matter on a decadal scale besides probable layered bubble trapping. Melt layers are untraceable prior to 2 ka BP due to the annual layers thinning. The analysis was performed on the ReCAP record cleaned from the possible melting- and the CFA technique-related spikes. This study leads, however, to a more detailed evaluation of the interpolar difference in the future work as the absolute value remains unresolved though we are confident that there were no big variations. We argue that the centennial variability in the CH4 is explained by the intertropical convergence zone global teleconnection and its influence on the monsoon activity and thus the CH4 production by tropical wetlands. The duration of the periods in CH4 concentration evolves through the time, which could potentially suggest a change in the atmospheric residence time of CH4.

This research investigated the potential use and application of radar, satellite, and other tracking data for sea ice and weather conditions in maritime-related Search and Research (SAR) operations in the Arctic. Specifically, this study analyzed a SAR event for a missing small vessel due into Utqiaġvik (formerly, Barrow), Alaska in July 2017 as well as archival records of U.S. Coast Guard SAR incidents in Arctic Alaska between 1976 and present. This study feeds into the Arctic Domain Awareness Center (ADAC) funded project - Developing sea ice and weather forecasting tools to improve situational awareness and crisis response in the Arctic - which seeks to create a prototype early warning sea ice and weather forecasting module for hazard planning in Utqiaġvik. This research found that data availability and accessibility, particularly in low bandwidth and further offshore areas, are challenges to data uptake during a SAR event. Nonetheless, the specific SAR incident in Utqiaġvik helped to illuminate there is a breadth of tools that can be applied and used in a SAR context - traditional and knowledge, modeling data, and USCG operational data. Specifically, modeling data from tools developed by the Arctic Domain Awareness Center (ADAC), the University of Alaska Fairbanks (UAF) and other research institutions was generated during this event to help support the local SAR effort. However, a level of pre- or post-processing was necessary in many cases, which can be a challenge for when data is needed immediately. This research holds implications for future use and uptake of modeling data in local SAR operations in Arctic Alaska and potentially the Arctic overall. Given that local SAR operators are predominantly the first line of response to maritime emergencies in Arctic Alaska, the ability to share and provide a set of resources to support SAR operators can be beneficial, particularly in a rapidly changing Arctic. A more targeted and systematic way to utilize and draw upon scientific research for SAR operations can potentially support the local SAR community, especially when immediate information is necessary. In particular, leveraging different products to validate, interpolate, and extrapolate information against one another, can help create more comprehensive situational awareness, especially for further offshore SAR events.

Man-made terraces for agricultural purposes are a quite diffuse and ancient anthropogenic landscape modification in mountainous areas. The original slope alteration, obtained through a sequence of sub-vertical and sub-planar surfaces, represents a human interference with the geomorphic system, altering the original balance of geomorphological and geo-hydrological factors. Stone walls and soil formed by human activity have been artificially immobilized on the slopes and are available again to gravitative processes once in abandonment and may be subject to deep degradation in case of intense rain events. However, socio-economic conditions play often a crucial role in the abandonment of terraces, indirectly contributing to increase gully erosion and walls failure. The modification of the original slope profile, due to its regular geometry in respect to the typical more complex natural surface, is rather suitable to be detected through remote sensing, particularly LIDAR, as many authors have recently demonstrated. In the present research the attention has been focused on the assessment of terraces and of the volume of stones and soil that have been involved by human activity. The research area is among the most deeply modified by terraces in the Mediterranean area and internationally famous for this landscape anthropogenic alteration: The Cinque Terre in Italy is a National Park intensively visited by tourists all over the year. Then terraces represent an important economic asset that need to be preserved from degradation and collapse as partially occurred in 2011 after the intense rain event that caused flood, hundreds of landslides and consequently damage. During the 2011 event many terraced slopes have collapsed with significant loss of soil and stone walls: the research allowed to evaluate the lost volumes and to estimate the remaining ones in the Vernazza catchment.

After 35 years of fairly steady lava effusion from the Pu’u ‘O’o vent along Kilauea’s mid portion of the East Rift Zone, a drastic change in the volcano’s behavior was initiated in May 2018. A series of new fissures opened between May 3rd and May 9th along the eastern end of the subaerial ERZ ( >40 km from the summit caldera), feeding a few short lived, relatively small volume flows. The products from the early phase of the eruption were dominated by plagioclase and clinopyroxene phenocrysts, gabbroic glomerocrysts and other crystal clots likely picked up prior to eruption. The initial activity transitioned between May 12th and around May 28th to steadier fountaining, higher effusion rate activity, leading to the formation of larger volume lava flows and the ensuing destruction of hundreds of houses and structures within the Puna area. This period marked the arrival of olivine-bearing Pu’u ‘O’o-like magma to the eruption site. Since the end of May, activity has focused on a unique vent (Fissure 8). This contribution examines the crystal cargo from the different magmas involved in the initial four weeks of the eruption, and their history through chemical zoning patterns of crystals. We discuss the magmatic origin of mineral assemblages from the evolved basalt end-member, and present initial results on timescales of magma interactions prior to their arrival at the surface from diffusion modeling in olivine.

Many biases in global climate models (GCMs) have been associated with the poor representation of unresolved variability due to organised convection in the tropics by the underlying cumulus parameterizations (CP). Many researchers have suggested that the quasi-equilibrium assumption (QEA) on which these CP’s are based is to blame. This is even more problematic with the recent and future increases in grid resolutions as the cloud large ensemble requirement for QEA breaks down. The stochastic multi-cloud model (SMCM) of Khouider et al. 2010 was proposed as a cheap alternative of overcoming this QEA dilemma by emulating the variability of the three cloud types which characterize tropical convection via a Markov jump birth-death process. The SMCM has proven to be very successful in terms of the simulation of the main modes of tropical variability when used as a simple alternative CP in a GCM. Here, we propose to incorporate the SMCM directly into the Zhang-McFarlane scheme (ZMS; Zhang and McFarlane 1995) to break the QEA dead end by using instead a stochastic plume ensemble and generalise the SMCM framework to cumulus schemes. The new stochastic ZMS (SZMS) uses a random number of plumes that are launched for each one of the three cloud types, shallow, congestus and deep, and that detrain at random levels, according to the SMCM. The new approach somehow combines the idea of Cohen and Craig (2006) of assuming a Poisson process for the number of plumes and that of Gentine et al. (2013) of prescribing a distribution of plume detrainment levels. Here we shall show the results of our experiment, for the single column version of the Community Earth System Model. Bibliography: Cohen BG, Craig GC. Fluctuations in an equilibrium convective ensemble. Part II: Numerical experiments. J. Atmos. Sci. 2006;63(8) Gentine P, Betts AK, Lintner BR, Findell KL, Van Heerwaarden CC, D’andrea F. A probabilistic bulk model of coupled mixed layer and convection. Part II: Shallow convection case. J. Atmos. Sci. 2013;70(6) Khouider B, Biello J, Majda AJ. A stochastic multicloud model for tropical convection. Comm. Math. Sci. 2010;8(1) Zhang GJ, McFarlane NA. Sensitivity of climate simulations to the parameterization of cumulus convection in the Canadian Climate Centre general circulation model. Atmos-ocean. 1995;33(3)

We employ a multi scale approach combining experiments and modelling to elucidate the impacts of small scale (sub-seismic resolution <10m) capillary pressure heterogeneities in field scale CO2 flow and trapping. We analyse 48 rock cores (~3cm length, 4cm diameter) covering the entire 100m interval of the Captain D sandstone in the Goldeneye field, UK North Sea. We experimentally measure porosity, capillary pressure, absolute permeability, relative permeability and trapping characteristics for the cores, which are used to create 3D numerical models with heterogeneities defined at the mm scale. These models are validated by predicting experimental X-Ray CT observations of saturation (mm scale) and pressure measurements (cm scale) at various flow rates and fractional flows of gas-water. The intrinsic, core scale properties are then used to populate 2D meso scale numerical simulations (50m x 10m size), with heterogeneities defined at cm scale using a geostatistical representation. We vary the correlation length and variance of the fields within the bounds of the experimental observations to investigate the impacts of small scale heterogeneity on vertical and lateral CO2 plume migration under different Capillary, Bond and Gravity numbers. At low flow potential, layered capillary pressure heterogeneities can speed up lateral plume migration by up to 20%, with gravitational segregation significantly enhancing the migration (See Fig. 1). In the vertical case, layered heterogeneities can significantly increase CO2 trapping, which is further enhanced with the inclusion of capillary pressure hysteresis. Finally, we derive capillary limit, upscaled equivalent properties from the meso scale simulations, which incorporate the impacts of small scale heterogeneity. These are used to represent the grid block properties in a full field 3D numerical model of the Goldeneye field (lateral ~5 km, depth 200m). We analyse the impact of varying relative permeabilities (with anisotropy) on the field scale plume migration and trapping. We see large differences compared to cases using intrinsic rock properties, indicating that proper inclusion of small scale heterogeneity is needed in upscaling workflows when predicting and assessing uncertainty in low flow potential CO2 plume migration at the field scale.

Since 2001, the Mars Exploration Program Analysis Group (MEPAG) has maintained a document outlining community consensus priorities for scientific goals, objectives, and investigations for the robotic and human exploration of Mars [1]. This “Goals Document” is a living document that is revised regularly (~every few years) in light of new Mars science results. It is organized into a hierarchy of goals, objectives, and investigations. The four Goals are not prioritized and are organized around major areas of scientific knowledge: “Life”, “Climate”, “Geology”, and “Preparation for Human Exploration”. Don Banfield is the current MEPAG Goals Committee Chair, and he oversees 2-3 representatives per Goal [2]. The most recent round of revisions (2018) was prompted by discussion at the 6th International Mars Polar Science and Exploration Conference (held in 2016 in Reykjavik, Iceland [3]), which pointed out that current high-priority Polar Science and Present-Day Activity questions were not well represented in content or priorities within the 2015 Goals Document. Upon request from the MEPAG Executive and Goals Committees [2], specific areas of disconnect were highlighted by representatives of the Mars Polar Science community; these were evaluated by the Goals Committee who proposed changes at sub-objective and investigation levels within the Climate and Geology Goals. These proposed changes were open for comment by the larger Mars community for 6 weeks, and then finalized. The official MEPAG 2018 Goals Document will be presented at the meeting. Additionally, the presentation will describe plans for the next round of revisions, which are expected to primarily come out of the presentations and discussion at the 9th International Conference on Mars (to be held at Caltech, Pasadena, CA in July 2019 [4]), and which are expected to include reference to returned sample science. The 2019 MEPAG Goals Document will form an important input to the next Planetary Science Decadal Survey [5]. [1] https://mepag.jpl.nasa.gov/reports.cfm?expand=science [2] https://mepag.jpl.nasa.gov/about.cfm [3] https://www.hou.usra.edu/meetings/marspolar2016/ [4] https://www.hou.usra.edu/meetings/ninthmars2019/ [5] NASEM, 2017. CAPS: Getting Ready for the Next Planetary Science Decadal Survey. https://doi.org/10.17226/24843.

An improved understanding of the mechanisms and factors affecting glacial flow is crucial to better predict sea level rise. Glacial ice often contains impurities such as the presence of small insoluble particles. Mixtures of ice and dust can be found in many places throughout the world, specifically in areas of high latitude and altitude (Moore, 2014). This study aims to understand the effect of entrained insoluble debris on processes of glacial motion. Glaciers move through a combination of internal ice deformation and basal sliding. Internal ice deformation, the flow of individual ice grains, has been found to be grain-size dependent in both field and laboratory studies (Goldsby and Kohlstedt, 2001). In an attempt to better understand ice grain size, this study considers the effect of debris on grain growth. Samples of pure ice and ice with debris were fabricated with a standard protocol and maintained at -5°C for controlled annealing. Microstructural characterization was preformed using a light microscope to image the samples, and calculating the average grain sizes using a linear-intercept method. The ice with debris was found to have smaller grain sizes, thought to be associated with grain-boundary pinning. Extrapolated values were used with a flow law, projecting that ice with debris will have lower viscosity, thus flow faster. To address basal sliding, the other form of glacial movement, we conducted a second phase of study. Basal sliding, the process of a glacier sliding over the bedrock, is influenced by the presence of meltwater at the base of the glacier (Hoffman et al., 2011). Frictional heating, from ice-on-rock friction, was studied as a factor affecting meltwater production. We conducted a simple 1D computer model using laboratory friction measurements of ice with entrained debris (Zoet et al., 2013). We find that debris content and frictional heating are directly proportional. Trials run at faster glacial velocities also show larger amounts of frictional heating. As frictional heating may increase meltwater, glaciers with debris may slide faster over bedrock. Overall, by better understanding the motion of debris-rich glaciers, we can focus our attention to areas around the world at risk, and better predict/prepare for sea level rise.

Our study uses field training data, airborne LiDAR (Light Detection and Ranging), imaging spectroscopy, and a Convolutional Neural Network (CNN) classifier to identify individual tree species in a mixed conifer forest in the Southern Sierra Nevada Mountains. The remote sensing data was collected on the National Ecological Observatory Network (NEON) Airborne Observation Platform in 2017. We trained the classifier using existing field plot data, and an independently collected validation dataset which identifies trees location of the 7 dominant species (Pine, Fir, Cedar and Oak), including condition and ‘live’ or ‘dead’ status. The LiDAR canopy height model was used to identify tree crowns and imaging spectroscopy data around these crowns were created as image ‘labels’. These species level ‘labels’ were used to train, test and validate a CNN tree species classifier. Our method achieved greater than 63-90% accuracy for all field validated stems and worked best for large diameter trees. On an independent tree stem dataset, we performed a species-level logistic regression to study which cases the classifier works most and least effectively. Spatially, in the southern areas scattered Black Oak were present and tree species often were confused with shrub species or covered by adjacent conifer species. In the north where the upper elevation forest is dominated by red and white fir the classifier achieved greater than 96% accuracy for larger canopy trees, with accuracy degrading to about 59-70% when smaller trees are included in the model. There is also genus level mis-classification, particularly between Red and White Fir species. There was high tree mortality in this forest and the classifier was effective in detecting large tree mortality, which also varies as a function of species and size of trees. This work leverages newly developed ‘deep learning’ tools which have yet to be extensively applied to the remote detection of large trees or in plant biogeography generally. This research is a proof-of-concept for forest community ecologists who want accurate ‘tree species maps’ to study how plants are distributed across space. Generally, this method will be of interest to biogeographers or remote sensing scientists looking to apply novel classification methods to problems beyond remote tree species identification.

We are investigating terrain deformations of long range and persistence that take place on continental scale over the stable shield of continental South America. The crustal deformation is being investigated by correlating with velocities and directions obtained from time series of approximately 60 GNSS stations deployed in Brazil and neighboring countries that take part of the Continuous network of the Geocentric Reference System for the Americas (SIRGAS-C). We attempt to estimate velocities and mid-plate strain rates using the best GNSS stations located in the stable South America mid-plate. The velocities have been estimated by Least Square Estimation (LSE) using SIRGAS weekly time series with a stochastic model composed of white noise plus flicker or random walk noises. Variance Component Estimation (VCE; Amiri-Simkooei et al., 2007) has been applied to classify the type of noise and compose the time series stochastic model. In addition, the time series breaks and offsets are taken into account in the LSE. The noises were classified as white plus flicker noise in approximately 70% of the horizontal component and most of random walk appears for stations located in the Amazon and Pantanal basins or near the coastal zones. The estimated formal precision reach about 0.10 mm/year with RMS of residual near 1.5 and 4.5 mm respectively for horizontal and vertical velocities. The estimated velocities by LSE were also computed by using the MIDAs code (Blewit et al 2015) and the results show an agreement of the order of 0.30 mm/y. The computed strain rates in the central part of Brazil indicate shortening, consistent with the predominance of reverse faulting mechanisms. When a station pair includes one station near the coast, the linear strain rates indicate extension. However, strain rates do not clearly indicate a preferred principal direction and do not seem compatible with the stress patterns derived from the focal mechanisms. In addition, the rate of seismic moment release indicates strain rates from earthquake occurrence two to three orders of magnitude lower than the observed strain rates from the geodetic network. Assessment of South America linear strain rates computed through estimated GNSS velocities will be discussed.