Understanding what causes weather-related stresses that lead to crop failures is a critical step towards stabilizing global food production. While there are many sources of weather-related stresses, the 30-60 day Madden-Julian Oscillation (MJO) is the dominant source of subseasonal climate variability in the tropics, making it a potential – but as of yet unexplored – source of crop failures. Here crop models and observational yield statistics are used to assess whether the MJO affects maize yields. We find that the influence of the MJO is widespread, affecting crop yields throughout the tropics. In dry, hot environments the MJO can lead to crop failures by reducing precipitation, decreasing soil moisture, and increasing extreme heat, while in wetter, cooler environments - where water stress is less common - MJO-forced decreases in rainfall bring increases in solar radiation that benefits crop yields. These results provide a pathway to develop actionable early warnings using subseasonal forecasts.

Results from a study of automatic aurora classification using machine learning techniques are presented. The aurora is the manifestation of physical phenomena in the ionosphere-magnetosphere environment. Automatic classification of of auroral images from the Arctic and Antarctic is therefore an attractive tool for developing auroral statistics and for supporting scientists to study auroral images in an objective, organized and repeatable manner. Although previous studies have presented tools for detecting aurora, there has been a lack of tools for classifying aurora into subclasses with a high precision (>90%). This work considers seven auroral subclasses; breakup, colored, arcs-bands, discrete, patchy, edge and clear-faint. Five different deep neural network architectures have been tested along with the well known classification algorithms; k nearest neighbor (KNN) and a support vector machine (SVM). A set of clean nighttime color auroral images, without ambiguous auroral forms, moonlight, twilight, clouds etc., were used for training and testing. The deep neural networks generally outperformed the KNN and SVM methods, and the ResNet-50 architecture achieved the highest performance with an average classification precision of 92%. Although the results indicate that high precision aurora classification is an attainable objective using deep neural networks, it is stressed that a common consensus of the auroral morphology and the criteria for each class needs to be established before classification of ambiguous images can be readily achieved.

The distribution of water in the upper mantle is believed to have strong influence upon global dynamics by influencing mantle rheology, modal mineralogy, melting systematics and chemical differentiation. The principal input of water is the process of subduction which is estimated to have introduced several ocean volumes to the mantle over Earth history. In principal, a large proportion of this water may be dissolved in the nominally anhydrous silicate minerals (NAMs), but quantifying this has been challenging. Xenoliths only sample the upper 200 km, less mostly, and suffer concern about alteration on the way to emplacement. Recent laboratory data show that seismic velocity is not sensitive to intracrystalline hydration. However, electrical conductivity is strongly sensitive and could provide estimates of mineral water content with suitable constraints. An 1300 km E-W transect of ~400 magnetotelluric (MT) soundings spanning the period range 0.01 to 17,480 s has been acquired from the northern California coast over the Gorda plate, across the Great Basin of Nevada and western Utah, and spanning most of the Colorado Plateau of eastern Utah. Regularized 2D inversion reveals an upper mantle whose resistivity below the broad Great Basin falls progressively with depth from values of ~100 ohm-m near 50 km to <10 ohm-m by 400 km depth. We test the hypothesis that the vertical resistivity profile is consistent with the maximal hydration degree allowed by ambient T-P short of triggering H2O-undersaturated melting (cf. Ardia, 2012, EPSL). An obvious possible source of hydration would be the Gorda, and to some extent the prior Farallon, subducting plates. Assuming standard and enhanced adiabats, deep resistivity profiles predicted using lab data of Novella (2017, Sci Rpts) suggest only resistivities in the near ‘hanging wall’ of the Gorda subduction zone under northwestern Nevada are low enough to represent full NAMs hydration. Under the central (eastern Nevada) and eastern (western Utah) Great Basin, large-scale resistivities are 2-3x too high, nominally. However, channelization of fluid upward from the plate could mean a mixed saturated-unsaturated peridotite upper mantle. Support has been from U.S. Dept of Energy contract DE-0006732 and National Science Foundation grants EAR-0838043 and OPP-1443532, and numerous prior.

Streams and rivers are important components of the carbon cycle as they simultaneously transport and transform terrigenous dissolved organic matter (DOM). The time DOM spends in a stream network is an important constraint on the biogeochemical processes that act upon DOM. We used high-resolution Fourier-Transform Ion Cyclotron Resonance Mass Spectrometry (FT-ICR MS) to study the spatial distribution of DOM at the molecular level at more than 100 sites within an Alpine stream network during summer and winter baseflow. We developed a model approximating the time DOM spent in the fluvial network. Discharge-weighted travel time (DWTT) explained the compositional changes of DOM, which differed markedly in summer and winter. We attribute the seasonal differences to differences in source material. Hydrological mixing at confluences was an important driver of the spatial dynamics of DOM. From the spatial patterns of individual DOM compounds we inferred the distribution of sources within the catchment, which differed seasonally. Finally, we estimated the apparent mass transfer coefficients of individual DOM compounds at the network level and identified the oxidative state of DOM as an important factor explaining uptake efficiency. This work contributes to our understanding of the spatial processes, temporal constraints and chemical properties of DOM that regulate the transformation and diagenesis of DOM at the fluvial network scale.

Near the inner edge of the plasma sheet, where the geomagnetic field transitions from dipolar to tail-like, very low values of the northward component of the field (Bz) are known to be occasionally exhibited, particularly in the substorm growth phase. It has been suggested that this may be a signature of a localized magnetic field dip, which are notoriously difficult to observe in situ. The existence of these localized minima is significant as they would be ballooning-interchange (BI) unstable. Previous work has investigated BI instability using localized particle-in-cell simulations with an imposed Bz minimum as an initial condition. However, evidence of the existence of localized Bz minima and BI instability at their tailward edges has been very limited in self-consistent global magnetosphere simulations. In this presentation, we demonstrate that the elusive nature of the instability has been due to the insufficient resolution of previous simulations. We present a highly-resolved global magnetosphere simulation, using our newly developed code Gamera. In a synthetic substorm simulation we demonstrate the formation of a Bz minimum localized in radius, 8-10 Re from Earth. The region becomes BI unstable in the substorm growth phase, leading to the formation of earthward and azimuthally propagating bubbles, distinct from those that form further downtail and become bursty bulk flows. These bubbles generate field-aligned currents and optical auroral signatures, similar to those observed on the ground and from space. We discuss the physical mechanisms for the formation of the localized Bz minimum by magnetic flux depletion, analyze the nature of the instability, characterize both magnetospheric and ionospheric signatures of the unstable region, and compare them with those observed.

The Ocean Observatories Initiative Cabled Array (OOI-CA) is a 25-year facility that provides unprecedented power and communications at Axial Seamount and sites on the Cascadia continental margin. Although the OOI is sometimes viewed as too expensive, the CA provides two capabilities that would be infeasible in autonomous configurations. First, at the expense of a reduced number of instrument sites, it enables reliable long-term observations with suites of standard sensors. Second, it can support novel instrumentation with exacting power, bandwidth and real time requirements. At the summit of Axial Seamount, the core sensor network spans the southern half of the caldera, and comprises 7 seismometers, 2 hydrophones, 4 bottom pressure and tilt instruments, and a variety of sensors in two hydrothermal fields. Additional sensors have also been added for testing. The volcano erupted in April 2015, within months of the CA coming online, providing new insights into the workings of caldera ring faults and support for the inference that eruptions at Axial occur at predictable levels of inflation. As of Summer 2019, the volcano has recovered 70% of the deflation that occurred in 2015. Another eruption is expected in a few years and several more are quite possible over the lifetime of the system. Because Axial has a shallow magmatic system that is well imaged, it is an excellent setting to study the links between volcanism and caldera dynamics and search for signals that are precursory to eruptions at a variety of timescales. Such studies would benefit from temporary deployments of autonomous sensors timed to coincide with the predicted times of eruptions. The CA should also be used to address questions related to the formation of hydrothermal event plumes and their role in flushing out fluids and microbes from the subsurface. This will require the addition of one or more cabled moorings above the caldera and enhanced sampling capabilities. There is a proposal to install several instrumented boreholes at the summit of Axial Seamount that would employ the CA to support novel interactive microbial and hydrologic subseafloor experiments. The CA can also be used to test emerging technologies such as resident autonomous underwater vehicles and distributed acoustic sensing, which will provide new tools to address key scientific questions.

The present space-based solar observing facilities, such as the Helioseismic and Magnetic Imager (HMI) instrument aboard the Solar Dynamics Observatory (SDO) satellite, can obtain time series photospheric vector magnetograms of solar active regions with high spatial and temporal resolution. The time evolution of the photospheric vector magnetic fields can be traced continuously for nearly all active regions appearing on the solar disk except for the marginal area near the solar limb. This facilitating condition is a huge benefit to the solar flare forecasting. By numerical modeling of the coronal magnetic fields based on the sophisticated nonlinear force-free field model from the observed photospheric vector magnetograms, the time series data of the coronal magnetic fields corresponding to the time series photospheric vector magnetograms can also be obtained. Then, the numerical analysis can be performed on the coronal magnetic field data, and the time evolution of the internal coronal magnetic structures as well as the evolution of the nonpotentiality in solar active regions can be revealed. The deduced quantitative measures of the coronal magnetic fields from the numerical analysis, such as the electric current density, the force-free factor, and the magnetic energy density, can be utilized in the solar flare forecasting and for establishing the solar flare prediction model. Samples of electric current analysis are presented. Through the numerical analysis of the time series magnetic fields in solar active regions, it is expected to predict flare locations, flare classes, and timing of flares.

We report on a feasibility study for an offshore instrument network in the Cascadia subduction zone to improve earthquake and tsunami early warning. The global DART buoy network provides effective warning for far-field tsunamis but near-field tsunami warning is challenging because the lead time is short and near-source observations are rarely available to directly measure the sea surface disturbance and evolution. Near-field tsunami warnings presently rely on rapid point source seismic inversions that do not estimate tsunami wave height. Efforts are underway to incorporate GNSS data into rapid source inversions that would support an initial near-field tsunami prediction. Offshore observations would contribute further to near-field tsunami warnings by providing: first, direct observations of seafloor and sea surface displacements during earthquake rupture and second, ongoing measurements for continued forecast refinement. Offshore instruments could also detect tsunamis triggered by submarine landslides and by so-called “tsunami” or “slow” or “silent” earthquakes that can generate unexpectedly large tsunamis but are characterized by shaking intensity so low as to be undetected or ignored. Pressure observations in the source zone will be challenging to interpret because they are dominated by seafloor accelerations and hydroacoustic waves rather than changes in hydrostatic pressure. In an effective system, pressure observations may need to be complemented by other observations such as inertial measurements of seafloor displacement, GNSS buoys and high-frequency coastal radar. It may also be important to place pressure sensors just seaward of the source zone to measure the developing tsunami in a region with an undisturbed seafloor. We will discuss alternative design options for an offshore instrument network in Cascadia, the research and development that must to be completed to determine the best approach, and the role of offshore observations in a holistic plan for tsunami mitigation.

SKB and several other waste management organisations have established the international SKB Task Force on Modelling of Groundwater Flow and Transport of Solutes (TF GWFTS) to support and interpret field experiments. An important objective of the task force is to develop, test and improve tools for conceptual understanding and simulating groundwater flow and transport of solutes in fractured rocks. Work is organised in collaborative modelling tasks. This study considers Task 9, which focuses on realistic modelling of coupled matrix diffusion and sorption in heterogeneous crystalline rock matrix at depth. This is done by inverse and predictive modelling of different in-situ transport experiments. The ultimate aim is to develop models that in a more realistic way represent retention in fractured rock. The Long-Term Diffusion and Sorption Experiment (LTDE-SD) was an in-situ radionuclide tracer test performed at the Äspö Hard Rock Laboratory at a depth of about 410 m below sea level. It is one of few recent in-situ studies focusing on tracer transport in the stagnant pore water of the crystalline rock matrix. The experimental results indicated a possible deeper penetration of tracers into the rock matrix than expected and the shape of the penetration profiles were not according to theory. Posiva’s REPRO (rock matrix REtention PROperties) experimental programme has been performed at the ONKALO rock characterisation facility in Finland. The two REPRO experiments considered were the Water Phase Diffusion Experiment, addressing matrix diffusion in gneiss around a single borehole interval, and the Through Diffusion Experiment, which is performed between sections of three boreholes. These three experiments provided an opportunity to improve the conceptual understanding of solute transport in fractured rock and to increase the realism in solute transport modelling, with the ultimate goal of improving safety assessments of deep geological disposal for nuclear waste. Of additional interest is the collective work performed by the task force to conceptually understand and interpret the field experiments, and at the same time increase the realism in solute transport modelling. This study would not have been possible without the support from the waste management organisations and the work by the multiple modelling teams.

Ganga river basin (GRB) in the Indian subcontinent is one of the most heavily irrigated land in the world. According to a book published in 2005 by Central Water Commission (CWC), 57% of the net irrigated land in India lies inside GRB only. Further GRB is also one of the most populous river basins in the world supporting almost 400 million people of India. With increasing use of fertilizers in agriculture and untreated sewage waste from the booming industries, there is need to assess the water quality and the contamination in surface water. We will use Soil Water Assessment Tool (SWAT) to model the hydrology of the river basin. For water quality analysis, SWAT is able to simulate the impact on hydrology, sediment and nutrients load, due to physical changes brought in the large ungauged river basins. We hypothesize that numerous small, rain-fed rivers in the Indo-Gangetic floodplain that are flowing predominantly through agricultural land are important non-point source of Nitrogen(N) and Phosphorus (P) and will control the nutrient budget of large river system. SWAT model will be used to simulate flow and nutrient/sediment concentrations of nitrogen/nitrates, phosphorus and sediment in the upper reach at Uttarkashi and Rishikesh, in the middle reach at Kanpur, Lucknow and Varanasi, and Farakka at the lower reach. SWAT model will be calibrated at daily/monthly time step for flow and monthly scale for water quality parameters. We will analyze the water quality in the basin using widely used Water Quality Index (WQI) considering pH, TDS, BOD, COD, hardness, nitrates, carbonates and silicates. We will use gridded climate data from Indian Meteorological Department (IMD) and water quality data from CWC. SRTM 90 m DEM, 300 m Land use/land cover map from Climate Change Initiative (CCI) and 7 km soil map from Food and Agriculture Organization (FAO).

The nature of the relationship between soil-moisture and subsequent precipitation is debated. Past studies have indicated that it varies by location, but there have been disagreements on feedback sign, strength and statistical significance over the same locations even while using the same data sets due to differences in statistical approaches. Here we re-evaluate estimates of feedback strength using Granger causality (following Tuttle and Salvucci 20016) for SMOS and AMSR-2soil moisture products and find that there is evidence both of there being a positive feedback over the entire nation (SMOS) and of there being a west-to-east gradient of positive to negative values (AMSR-2, similar to what was found by Tuttle). This project has demonstrated that differences arise using identical statistical estimation techniques with different satellite estimates of soil moisture. The passive microwave AMSR-E and AMSR-2 data produce a gradient of positive to negative impact factors across the U.S. while the SMOS data set yields a mostly positive impact factor over the entirety of the U.S. Here we explain this discrepancy by noting that the areas where the sign of the impact differ are areas where the soil moisture perturbations of SMOS and AMSR-2 are anticorrelated, and also that these areas coincide with areas of high vegetation cover, which are well known to challenge remote sensing estimates of moisture.

Key points: • The detrimental impact of preadsorbed water on the methane adsorption capacity and rate is more pronounced than that of water vapor • The distributions of adsorbed methane and water in shale pores were compared between the SAWM and PMMS • Water vapor preferentially enters mesopores (1.5-20 nm), and preadsorbed water mainly occupies micropores (0.3-1.5 nm) Abstract Water plays an essential role in shale gas migration and adsorption, and most studies on the influence of water on shale gas adsorption refer only to moisture-equilibrated shales. To investigate the impact of water vapor on methane adsorption in shales, three experiments were conducted and compared: (1) pure methane adsorption onto dry shale (PMD), (2) pure methane adsorption onto moisture-equilibrated shale (PMMS), and (3) simultaneous adsorption of water vapor and methane (SAWM) onto shale. Comparison of the experimental results reveals that the detrimental impact of water vapor on methane adsorption is inferior to that of preadsorbed water. Two mechanisms, i.e., water blocking and adsorption competition, are responsible for the reduction and difference in the methane adsorption capacity and adsorption rate between the PMMS and SAWM. Compared to the PMD, the methane adsorption capacity decreases by 81-96% in the PMMS, and by 20-49% in the SAWM. Methane adsorption equilibrium is achieved the fastest in the PMD. Before the equilibration degree reaches 95%, methane adsorption during the SAWM progresses more rapidly, while the reverse occurs when the equilibration degree exceeds 95%. The pore size distribution and water film thickness calculations indicate that the impacts of water vapor in the SAWM on micro-to mesopores are weaker than those of preadsorbed water. In the PMMS, adsorbed water mainly

Accurate forecast of solar irradiance remains a major challenge, especially under the influences of aerosols, clouds and aerosol-cloud interactions due to their inadequate parameterizations in numerical prediction models. This study focuses on the impacts of cloud microphysics and the indirect aerosol effect on solar irradiance. The state of art Weather Research and Forecasting model specifically designed for simulating and forecasting solar radiation (WRF-Solar) is employed to investigate the sensitivity of the total solar irradiance and its partitioning into direct and diffuse irradiances to aerosol and cloud properties. First, a number of microphysical schemes will be tested against the measurements of shallow cumulus and stratiform clouds at the DOE ARM SGP site. Efforts will be made to quantify the uncertainty spread. The effects of cloud microphysics on surface solar irradiance will be identified. Second, the indirect aerosol effect on cloud formation and thus surface solar irradiance will be investigated by using the Thompson aerosol aware microphysical scheme and different treatment of aerosols. In particular, we will examine the aerosol indirect effects in different cloud regimes. To address the aforementioned problems, we will introduce a new model evaluation framework based on different WRF-Solar setups (nested WRF, WRF-LES, and single column WRF). In addition, different evaluation metrics will be used, including the RMSE, MAPE, and relative Euclidean distance. The results will provide physical insight into the understanding of aerosol-cloud-radiation interactions and into improving solar radiation forecast in cloudy conditions.

Redox input by subducting slab into deep mantle is of vital importance for deep cycle and isotopic evolution of volatile elements, whose chemically stable forms are controlled by redox state. Lithospheric mantle is crucial in redefining redox state of the Earth’s deep mantle. To constrain to which extent subducted slab can modify redox state of the upper mantle and how much oxygen slab can carry into deep Earth, we investigated redox kinetics of olivine adopting diffusion couple methods at 1 GPa and 1373-1573 K in a piston cylinder apparatus. It is found that redox process in olivine is diffusion-controlled, and diffusing on the order of 10-12 m2/s at 1473 K. The oxidation process in initially reduced olivine is oxygen fugacity (fO2)-independent with activation enthalpy of 235±56 kJ/mol, while the reduction process in initially oxidized olivine is fO2-dependent with an exponent of 2/5. Diffusion profile analysis reveals that redox state of starting material plays decisive role in determining redox mechanism. Below ΔFMQ+1, redox process in olivine is controlled by oxygen grain boundary diffusion, while above ΔFMQ+1, it is rate-limited by faster diffusion species which might be hydrogen related Mg vacancy. The extremely slow redox rate limits the homogenization of the slab and its surrounding mantle as redox state of the upper mantle remains unchanged for over past 3.5 Gyrs. The subducted slab has the ability to efficiently transport oxidized components to the region deeper than the mantle transition zone. A highly underestimated oxygen reservoir may have formed in the deep Earth.

The style of convective force transmission to plates and strain-localization within and underneath plate boundaries remain debated. To address some of the related issues, we analyze a range of deformation indicators in southern California from the surface to the asthenosphere. Present-day surface strain rates can be inferred from geodesy. At seismogenic crustal depths, stress can be inferred from focal mechanisms and splitting of shear waves from local earthquakes via crack-dependent seismic velocities. At larger depths, constraints on rock fabrics are obtained from receiver function anisotropy, Pn and P tomography, surface wave tomography, and splitting of SKS and other teleseismic core phases. We construct a synthesis of deformation-related observations focusing on quantitative comparisons of deformation style. We find consistency with roughly N-S compression and E-W extension near the surface and in the asthenospheric mantle. However, all lithospheric anisotropy indicators show deviations from this pattern. Pn fast axes and dipping foliations from receiver functions are fault-parallel with no localization to fault traces and match post-Farallon block rotations in the Western Transverse Ranges. Local shear wave splitting inferences deviate from the stress orientations inferred from focal mechanisms in significant portions of the area. We interpret these observations as an indication that lithospheric fabric, developed during Farallon subduction and subsequent extension, has not been completely reset by present-day transform motion and may influence the current deformation behavior. This provides a new perspective on the timescales of deformation memory and lithosphere-asthenosphere interactions.

The spectacular fossil plant assemblage preserved in the non-marine Mount Flora Formation of the northern Antarctic Peninsula represents the diverse Jurassic flora that once covered the Gondwanan continents at high paleolatitudes. The depositional facies of the formation plays a key role in the tectonic interpretations and basin evolution models that attempt to reconcile large igneous province magmatism, continental break up, and magmatic arc development throughout the Jurassic and Cretaceous. Limited U-Pb in situ geochronology reported from the Mount Flora Formation and adjacent units lack the necessary resolution to overcome ambiguous correlations and biostratigraphic discrepancies. We present two high-precision U-Pb zircon ages (CA-ID-TIMS method) from a distinct tuffaceous interval of the Mount Flora Formation at Hope Bay, which document a terminal Middle Jurassic age (Callovian Stage) for the formation and its paleoflora. In excess of 1400 new collected fossil plant specimens exhibit a highly diverse Jurassic plant association that dominated the Antarctic Peninsula nearly 17 million years after its disappearance from northern Patagonia. This suggests similar paleoecological conditions were established diachronously throughout basins of southern Gondwana, possibly facilitating floral migrations in response to local climate change. The depositional facies of the Mount Flora Formation, its age proximity to the marine Nordenskjöld Formation in the Antarctic Larsen Basin, and its coincidence with a regional unconformity in the northern Patagonia point out to a complex interplay among magmatic arc development, tectonic extension and continental break up that dominated the geologic and paleoenvironmental evolution of southern Gondwana near the end of the Middle Jurassic.

Regional water storage monitoring is very essential and critical for regional water resource management, hydro-ecological sustainable development and studying interaction between hydrology and climate. We used multi-mission satellite dataset, a global hydrological model, and ground meteorological data to investigate the terrestrial water storage changes (TWSc) in the Qaidam basin (QB). Terrestrial water storage (TWS) exhibited an increasing trend for the past 17 years, and TWSc obtained from different methods or dataset were consistent with precipitation. Through the singular spectrum analysis, we obtained the interannual and seasonal components of TWSc, and significant shifts in the interannual TWSc were observed in the QB. TWS exhibited an increase from 2002 to 2013, a decrease from 2013 to 2016, and an accelerated increase after 2016. Additionally, the study reported that the interannual change in TWS was strongly correlated with the Pacific Decadal Oscillation (PDO), compared to Arctic Oscillation and El Nino-Southern Oscillation.

Accurate fire emissions inventories are crucial to predict the impacts of wildland fires on air quality and atmospheric composition. Two traditional approaches are widely used to calculate fire emissions: a satellite-based top-down approach and a fuels-based bottom-up approach. However, these methods often considerably disagree on the amount of particulate mass emitted from fires. Previously available observational datasets tended to be sparse, and lacked the statistics needed to resolve these methodological discrepancies. Here, we leverage the extensive and comprehensive airborne in situ and remote sensing measurements of smoke plumes from the recent Fire Influence on Regional to Global Environments and Air Quality (FIREX-AQ) campaign to statistically assess the skill of the two traditional approaches. We use detailed campaign observations to calculate and compare emission rates at an exceptionally high resolution using three separate approaches: top-down, bottom-up, and a novel approach based entirely on integrated airborne in situ measurements. We then compute the daily average of these high-resolution estimates and compare with estimates from lower resolution, global top-down and bottom-up inventories. We uncover strong, linear relationships between all of the high-resolution emission rate estimates in aggregate, however no single approach is capable of capturing the emission characteristics of every fire. Global inventory emission rate estimates exhibited weaker correlations with the high-resolution approaches and displayed evidence of systematic bias. The disparity between the low resolution global inventories and the high resolution approaches is likely caused by high levels of uncertainty in essential variables used in bottom-up inventories and imperfect assumptions in top-down inventories.