One of the next giant leaps for humanity—inhabiting our neighbor planet Mars—requires enough water to support multi-year human survival and to create rocket fuel for the nearly 150-million-mile return trip to Earth. Water that is already on Mars, in the form of ice, is one of the leading in situ resources being considered in preparation for human exploration. Human missions will need to land in locations with relatively warm temperatures and consistent sunlight. But in these locations, ice (if present) is buried underground. Much of the ice known to exist in mid-latitude locations was likely emplaced under climate conditions (and orbital parameters) different from today. So in addition to providing an in-situ resource for human exploration, Martian ice also provides a crucial record of planetary climate change and the effects of orbital forcing.This presentation will highlight techniques and recent activities to characterize Mars’ underground ice, such as the Subsurface Water Ice Mapping (SWIM) Project (Morgan et al. 2021, Nature Astro.; Putzig et al. In Press, Handbook of Space Resources; Putzig et al. this AGU; Morgan et al. this AGU). We present outstanding questions that will be vital to address in the context of ISRU (in situ resource utilization) and connections between these questions and the climate in which the ice was emplaced and evolved (e.g., Bramson et al. 2020, Decadal White Paper). Lastly, we discuss how these science activities intersect with future exploration, particularly that enabled by collaborations between space agencies as well as industry partners (Heldmann et al. 2020, Decadal White Paper; Golombek et al. 2021, LPSC).High-priority future work includes better orbital characterization of shallow ice deposits, such as radar sounding at shallower scales (<~10m) than that of SHARAD, as proposed for the International Mars Ice Mapper. Also needed are detailed studies of the engineering required to build potential settlements at specific candidate locations; this includes characterization of the nature of the overburden above the ice, which will inform future resource extraction technology development efforts. Ideally, initial landing sites would be chosen with a long-term vision which includes preparation and development of the basic technologies and designs needed for human landing on Mars.

Traditionally, walnut kernel color is scored by a human on a four-point color scale ranging from extra light to Amber, developed by the California DFA. Two important reasons for using a high throughput machine vision system for kernel phenotyping is to obtain better quantitative resolution to use in molecular breeding and to avoid errors in human vision phenotyping. RGB and LAB values give much more depth to qualitative calculations, and machine data is far more consistent than human scoring. Previous studies on walnut kernel imaging utilized a thresholding-based approach to segment walnut kernels from their background, however, detection accuracy can be improved. In this study, we make changes to detection methods, primarily using PyTorch computer vision processing and an improved thresholding method to better segment walnut kernels. The PyTorch CNN pipeline allows us to use specific photos to train a model, and the model can segment photos without the need to figure out image thresholds. Our proposed thresholding method uses the magick package in R instead of the ImageJ macros that were previously used. After taking photos with a CVS, we used the CNN model as well as an R script to identify and segment kernels from the background. Our preliminary data shows that with enough training, the CNN model is more robust in edge cases where we have overlapping kernels or shifted images. This paper will focus on exploring the differences between the three thresholding methods, and use the best method for future breeding projects.

The Vaisigano River which flows through the Apia capital of Samoa is in a characteristic short and steep catchment conducive to rapid flash flooding following intense periods of antecedent rainfall. This results in short early warnings and emergency response lead times. Through the Government of Samoa’s Vaisigano Catchment Project (VCP) supported by the Green Climate Fund, technological initiatives to improve the forecasting of imminent flooding in the catchment which enables longer early warnings and response lead times were undertaken within a hazard risk context. In this talk we describe a pilot impacts-based flood monitoring, early warnings, and decision support system developed through the VCP and tailored for the Vaisigano River. The system comprises an integrative real-time automated framework involving the ingestion of numerical weather prediction rain intensity forecasts, real-time rainfall, river level and flow monitoring data, precomputed rainfall-runoff and predictive flood peak and magnitude tools, as well as estimates of flood inundation exposure and threat to safety at buildings and on roads for different return period events. Information is ingested into a centralized, web-based, flood decision support system (FDSS) portal that enables hydrometeorological officers to monitor, forecast and alert relevant emergency or humanitarian responders of imminent flooding with adequate lead time. The FDSS was tested in the lead up to the 18 December 2020 flooding in the Vaisigano and was able to alert duty officers of the estimated timing and magnitude of imminent channel-overtopping with up to 24 hours lead time. We discuss some of the key challenges and gaps to guide system improvements, as well as offer recommendations for future work.

Scientific and related management challenges in the water domain require synthesis of data from multiple domains. Many data analysis tasks are difficult because datasets are large and complex; standard formats for data types are not always agreed upon nor mapped to an efficient structure for analysis; water scientists may lack training in methods needed to efficiently tackle large and complex datasets; and available tools can make it difficult to share, collaborate around, and reproduce scientific work. Overcoming these barriers to accessing, organizing, and preparing datasets for analyses will be an enabler for transforming scientific inquiries. Building on the HydroShare repository’s established cyberinfrastructure, we have advanced two packages for the Python language that make data loading, organization, and curation for analysis easier, reducing time spent in choosing appropriate data structures and writing code to ingest data. These packages enable automated retrieval of data from HydroShare and the USGS’s National Water Information System (NWIS), loading of data into performant structures keyed to specific scientific data types and that integrate with existing visualization, analysis, and data science capabilities available in Python, and then writing analysis results back to HydroShare for sharing and eventual publication. These capabilities reduce the technical burden for scientists associated with creating a computational environment for executing analyses by installing and maintaining the packages within CUAHSI’s HydroShare-linked JupyterHub server. HydroShare users can leverage these tools to build, share, and publish more reproducible scientific workflows. The HydroShare Python Client and USGS NWIS Data Retrieval packages can be installed within a Python environment on any computer running Microsoft Windows, Apple MacOS, or Linux from the Python Package Index using the PIP utility. They can also be used online via the CUAHSI JupyterHub server (https://jupyterhub.cuahsi.org/) or other Python notebook environments like Google Collaboratory (https://colab.research.google.com/). Source code, documentation, and examples for the software are freely available in GitHub at https://github.com/hydroshare/hsclient/ and https://github.com/USGS-python/dataretrieval.

In consultation with the Samoa Electric Power Corporation (EPC) and the Samoa Meteorology Division (SMD), the Afulilo Water Storage Outlook Module (AWSOM) that was developed as a manually operated spreadsheet application during COSPPac-1, has now been redeveloped as an automated web application (AWSOM-2). The Afulilo Hydropower Scheme is the largest renewable power scheme in Samoa and is central to Samoa’s goal of becoming 100% renewable in the energy sector by 2025. AWSOM-2 draws on weekly, monthly, and seasonal rainfall forecast products from the ACCESS-S forecasting system, as well as weather and climate forecasts from global models. Additionally, AWSOM-2 draws on rainfall observations from the dam, dam level measurements conducted by EPC and the Samoa Water Resource Division, and power generation rates being operated by EPC. The model incorporates physical relationships derived from studies of how the reservoir responds to rainfall, water runoff from the upper catchment, and losses from evapotranspiration and seepage. Samoan Met staff operationally review model outputs, add interpretive commentary from local knowledge and perspectives, and then forward the reservoir storage outlook report to EPC. This enables EPC to consider options for water use for power generation and optimise water use while maintaining a guaranteed electricity supply. AWSOM-2 has been coded in python by NIWA, through funding from the NZ Ministry of Foreign Affairs and Trade. The application is run on the CliDEsc server at SMD.

In spite of more than 20 years of substantial advances, solar flare prediction remains a largely outstanding problem. This is partly because of the scarcity of major flares. Effective flare prediction, if ever achieved, would help mitigate a substantial projected economic damage, with a long-range magnitude of 1 to 2 trillion dollars for the US alone. Prediction could also help mitigate, or even prevent, serious health risks to astronauts exposed to flares’ electromagnetic radiation and particulate. While many recent flare prediction studies have opted to employ Machine Learning techniques to better tackle the problem, a lack of sufficient understanding of how to properly treat the data often leads to overly optimistic results. We use the recently generated GSU solar flare benchmark dataset, called Space Weather ANalytics for Solar Flares (SWAN-SF), to show how a ‘mediocre’ forecast model can turn into an ‘impressive’ one, by simply overlooking some basic practices in data mining and machine learning. The benchmark is a multivariate time series collection, extracted from magnetographic measurements in the solar photosphere and spans over eight years of the Solar Dynamics Observatory Helioseismic and Magnetic Imager (SDO/HMI) era. We briefly explain the data collection process, the sampling and the slicing of time series, and then outline a series of experiments using machine learning models to illustrate the common mistakes, fallacies and pitfalls in forecasting rare events. We particularly elaborate on how and why imbalanced datasets, in general, impact the models’ performance, and how different under- or over-sampling methodologies and weighting practices could introduce accurate but often weak models. Concluding, we aim to draw attention to the impact of these practices on the flare forecasting models and how to train models by accentuating the statistical robustness over a relative accuracy in prediction.

Continuous collection and analysis of high-resolution phenotype data is critical to develop crops resilient to the consequences of climate change. Though web-accessible tools for parallel, reproducible scientiSic workSlows render big data increasingly tractable, software for plant science remains inadequate for large-scale precision agriculture. Cyberinfrastructure must present minimal barriers to entry, accommodate rapidly changing dependencies, support a wide variety of use cases, and weave together sensors at the edge, laptops, clusters, and cloud storage into a coherent virtual workspace. PlantIT is a web portal intended as such an environment. Platforms like PlantIT and its precursor DIRT [1] permit efSicient phenotyping and equip geographically distributed researchers with a code-optional interface. WorkSlows are published in Docker images, deployed as Singularity containers to public or private computing resources, and monitored in real time. Data are stored automatically in the CyVerse Data Store and can be annotated according to the MIAPPE [2] standard. GitHub integration provides versioning and repositories can be activated with a single conSiguration Sile, like Travis or GitHub Actions. Containers allow for a range of use cases, including image-based trait measurements, 3D reconstructions, morphological growth simulations, and crop modeling. Pseudo-batch/stream processing is also necessary; as data scales, manual batch jobs rapidly become infeasible, and (re-)analysis must occur upon arrival in near-real-time. We suggest web-accessible phenotyping automation software may address bottlenecks and help reveal undiscovered relationships between genes, traits, and the environment.

Robust characterization of rock anisotropy is the preferred laboratory method to support seismic data interpretation in the field. Especially in shale formations, accurate elastic anisotropy helps delineate subsurface stress distribution, improve seismic imaging, and enhance hydraulic fracturing design. The conventional technique for evaluating rock elastic anisotropy involves ultrasonic pulse transmission between source and receiver transducers attached to the rock surface. The size, position and orientation of the source and receiver in relation to the propagation distance and direction, and their coupling to the rock surface introduce undesired uncertainties in Thomsen’s anisotropy parameters and rock attenuation: effective propagation distance; group or phase velocity; impact of the contact interface on measured wave attenuation; impact of heterogeneity on wave velocity measurements. We apply here the contactless laser ultrasonic method, involving a source laser (short-pulse high-peak power), a probing laser (vibrometer), and a cylindrical rock sample set on a rotating stage. The footprint of the source laser beam is 2 mm, and that of the receiver beam is 0.1 mm, which can conveniently be approximated by a point on a centimetric rock sample. The propagation distance is hence unambiguously known, implying that a group velocity is effectively estimated, and the observed attenuation is solely due to the rock, not to the rock-transducer interface (extrinsic). The technique also allows for a denser ultrasonic probing. Four samples are probed, where the P-wave velocity along up to 630 independent ray paths is evaluated. Three samples are made of a known, homogeneous, and layered synthetic material phenolic grade, approximately transversely isotropic. These samples were cored along, across and at 45° to the layers. The fourth sample is a heterogeneous shale from the Goldwyer formation (Canning basin, Western Australia). The measurements on the three known phenolic samples are used to validate the method, and optimise the measurement protocol. Application of the method to the unknown heterogeneous shale suggests that (i) anisotropy can be reliably estimated in the homogeneous sub-volume of the sample and that (ii) the mineralogical heterogeneity can be detected and identified in other sub-volume.

“Eddy saturation” is the regime in which the total time-mean volume transport of an oceanic current is relatively insensitive to the wind stress forcing and is often invoked as a dynamical description of Southern Ocean circulation. We revisit the problem of eddy saturation using a primitive-equations model in an idealized channel setup with bathymetry. We apply only mechanical wind stress forcing; there is no diapycnal mixing or surface buoyancy forcing. Our main aim is to assess the relative importance of two mechanisms for producing eddy saturated states: (i) the commonly invoked baroclinic mechanism that involves the competition of sloping isopycnals and restratification by production of baroclinic eddies, and (ii) the barotropic mechanism, that involves production of eddies through lateral shear instabilities or through the interaction of the barotropic current with bathymetric features. Our results suggest that the barotropic flow-component plays a crucial role in determining the total volume transport.

Cities are the most sensitive and vulnerable places to climate variability and change and weather-related extreme events given high population density, with the aggravating factor that urban climate also suffers modifications due to the widespread replacement of the natural surface altering the local thermal conditions. The Aburrá Valley is a narrow valley located at the tropical Andes in northern South America with urban areas between 1300 and 2000 m.a.s.l, a population of approximate 3.9 million people, and a comfortable climate relative to standard indoor conditions. In this work, we examine observed weather patterns in the local scale and the urban canopy layer (UCL) using data from weather stations at sites with different surface features regarding vegetation/non-impervious fractions and urban structure (Sky View Factor SVF). UCL data is available from two data sources, the first one from a field campaign using all-in-one weather sensors at the valley´s bottom, and the second one from a low-cost sensor network with robust temperature and humidity data as part of a local citizen science project with measurements in a diverse altitude range. Results suggest that at the local scale there exist different climate mean conditions due to altitude, with significant weather variability depending on radiation levels and rainfall occurrence, but at the same time, the urban effects are evident since the lowest altitude stations do not necessarily register the highest temperatures depending on the local characteristics. UCL measurements show that, while the altitude defines a background state, there are notable differences between places mainly influenced by insolation changes due to vegetation around and above sensors. Currently, the local population does not perceive thermal stress as a risk factor because it is not difficult to find places with appropriate thermal conditions when thermal discomfort arises. However, this research is relevant considering the projected local surface temperature increase due to climate change and the inexistence of baseline studies assessing the thermal comfort in outdoors to support local adaptation actions. The results of this study are useful for urban planning and building design to improve thermal conditions, especially in open spaces.

On the 22nd of May 2021, although no alarming precursory unrest had been reported, Nyiragongo volcano erupted and lava flows threatened about 1 million of inhabitants living in the cities of Goma (Democratic Republic of Congo) and Giseny (Rwanda). After January 1977 and January 2002, it was the beginning of the third historically known flank eruption of Nyiragongo volcano and the first ever to be recorded by dense measurements both on the ground and from space. In the following days, seismic and geodetic data as well as fracture mapping revealed the gradual southward propagation of a shallow dike from the Nyiragongo edifice underlying below Goma airport on May 23-24, then Goma and Gisenyi city centers on May 25-26 and finally below the northern part of Lake Kivu on May 27. Southward migration of the associated seismic swarm slowed down between May 27 and June 02. Micro seismicity became more diffuse, progressively activating transverse tectonic structures previously identified in the whole Lake Kivu basin. Here we exploit ground based and remote sensing data as well as inversion and physics-based models to fully characterize the dike sized, the dynamics of dike propagation and its arrest against a structural lineament known as the Nyabihu Fault. This work highlights the shallow origin of the dike, the segmented dike propagation controlled by the interaction with pre-existing fracture networks and the incremental crater collapse associated with drainage which led to the disappearance of the world’s largest long-living lava lake on top of Nyiragongo.

A statistical-equilibrium, geostrophic-turbulence regime of the stochastically forced, one-layer, reduced-gravity, quasi-geostrophic model is identified in which the numerical solutions are representative of global mean, mid-latitude, open-ocean mesoscale variability. A nominal best fit to observed SSH variance, autocorrelation, eddy, and spectral statistics is obtained for dimensional SSH stochastic-forcing variance production rate 1/4 cm^2/d, an SSH damping rate 1/62 1/wk, and a stochastic forcing autocorrelation timescale near or greater than 1 wk. This ocean mesoscale regime is nonlinear and appears to fall near the stochastic limit, at which wave-mean interaction is just strong enough to begin to reduce the local mesoscale variance production, but is still weak relative to the overall nonlinearity. Comparison of linearly-inverted wavenumber-frequency spectra shows that a strong effect of nonlinearity, the removal of energy from the resonant linear wave field, is resolved by the gridded altimeter SSH data. These inversions further suggest a possible signature in the merged altimeter SSH dataset of signal propagation characteristics from the objective analysis procedure.

To address climate change, social issues need to be integrated in geoscience-related STEM research and curriculum. However, social issues are often ignored in STEM fields, even though STEM practices are usually implicitly rooted in inequitable values and practices. A previous study showed that some students do not see the relevance of diversity, equity, and inclusion (DEI) activities when they are presented as separate from technical content, and some students argued that social content was irrelevant or inappropriate in a technical STEM course. Some engineering faculty have identified the lack of curriculum as a barrier to integrating the socially embedded nature of STEM into their introductory courses. Therefore, we developed a case that addresses social issues such as race, class, and gender in relationship to the social and environmental characteristics of New Orleans and its built environment. The case shows how marginalized people, who have been historically and systemic discriminated against, suffered the greatest losses under the impacts of climate-change driven Hurricane Katrina. The case uses student-centered learning strategies, is applicable for use in a range of courses, including many geoscience courses, and will be publicly available. In Fall 2019 the case will be taught in two first year undergraduate civil and environmental engineering courses (n ≈120) at a R1 university. Coursework from the case will be analyzed using Qualitative Content Analysis, which distills responses into categories and themes that characterize both commonalities and differences. While we do not yet have results, results will be presented at AGU. Our results will provide insight into the effectiveness of our case study as an intervention as well as challenges and successes in implementation, which may help others implement this or other similar activities. More broadly, our intervention demonstrates one way of integrating content more commonly found in interdisciplinary courses into technical, disciplinary-focuses courses. Our study will provide insight into how students respond when DEI and technical content are interconnected, helping further our ability to prepare students in a range of geoscience-related courses to respond to the challenges of climate change through an equity and justice-oriented lens.

To study the role of the Atlantic meridional overturning circulation (AMOC) in climate change, we perform an abrupt CO2–doubling experiment using a coupled atmosphere-ocean- ice model with a simple geometry that separates the ocean into small and large basins. As in observations and high-end climate models, the small basin exhibits a MOC and warms at a faster rate than the large basin. In our set-up, this contrast in heat storage rates is 0.6 ± 0.1 W/m^2, and we argue that this is due to the small basin MOC. However, the MOC weakens significantly, yet this has little impact on the small basin’s heat storage rate. We find this is due to the effects of both compensating warming patterns and interbasin heat transports. Thus, although the presence of a MOC is important for enhanced heat storage, MOC weakening is surprisingly unimportant.

Oceanic basalts provide an invaluable window into evolutionary processes governing mantle spatial and temporal chemical heterogeneity. Ocean island basalts (OIBs) and enriched mid-ocean ridge basalts (E-MORBs) are powerful tracers of mantle melting and crust-mantle recycling processes. Whether the elemental and isotopic variations observed in both E-MORBs and OIBs are derived from similar mechanisms, however, remains under debate. Investigating compositional differences between E-MORBs and OIBs is a simple approach to constrain their origins, a technique for which machine learning classification algorithms are optimal. Here we implemented a novel machine learning approach complemented by mantle component mixing models to highlight compositional differences between E-MORBs and OIBs and further investigate their petrogenesis (data sourced from GEOROC database and Gale et al., 2013). Considering Random Forest-based Gini indexes, elements sensitive to pressure and degree of melting (FeO, TiO2, Lu, and Sr) were identified as the best discriminators between E-MORBs and OIBs. Our Gaussian process classification algorithm successfully classified OIBs and E-MORBs better than 97% of the time when considering 1) Sr & FeO and 2) TiO2 & Lu. The probabilistic nature of Gaussian process modeling permitted calculation of new quantitative discriminant diagrams rooted in probability (Sr vs. FeO and TiO2 vs. Lu). Complementary trace element modeling yielded compositionally similar E-MORB and OIB sources with moderately incompatible element enrichments in the OIB source due to the influence of recycled oceanic crust (Prytulak & Elliott, 2007). Our source compositions are consistent with a simple, joint model for E-MORB and OIB petrogenesis after Donnelley et al. (2014): low-degree partial melts of subducted slabs metasomatize the depleted mantle producing a re-fertilized mantle (RM). RM is randomly sampled at mid-ocean ridges to produce E-MORB, while upwelling plumes sample both RM and recycled oceanic crust, yielding OIB. References: Donnelly et al. (2004). Earth and Planet. Sci. Lett., 226(3–4), 347–366. Gale et al. (2013). Geochem., Geophys., Geosyst., 14(3), 489–518. Prytulak & Elliott (2007). Earth and Planet. Sci. Lett., 263(3–4), 388–403.

India receives more than 80% of annual rainfall during the summer monsoon season of June – September. Extreme rainfall during summer monsoon season causes severe floods in many parts of India, annually. The floods in Kerala in 2019; Chennai during 2015 and Uttarakhand in 2013 are some of the major floods in recent years. With high population density and weaker infrastructure, even moderate precipitation extremes result in substantial loss to life and property. Thus, understanding and modeling the return levels of extreme precipitation in space and time is crucial for disaster mitigation efforts. To this end, we develop a Bayesian hierarchical model to capture the space-time variability of –summer season 3-day maximum precipitation over India. In this framework, the data layer, the precipitation extreme – i.e., seasonal maximum precipitation, at each station in each year is modeled using a generalized extreme value (GEV) distribution with temporally varying parameters, which are decomposed as linear functions of covariates. The coefficients of the covariates, in the process layer, are spatially modeled with a Gaussian multivariate process which enables capturing the spatial structure of the rainfall extremes and covariates. Suitable priors are used for the spatial model hyperparameters to complete the Bayesian formulation. With the posterior distribution of spatial fields of the GEV parameters for each year, posterior distribution of the nonstationary space–time return levels of the precipitation extremes are obtained. Climate diagnostics will be performed on the 3-day maximum precipitation field to obtain robust covariates. The model is demonstrated by application to extreme summer precipitation at 357 stations from this region. Preliminary model validation indicates that our model captures historical variability at the stations very well. Maps of return levels provide spatial and temporal variability of the risk of extreme precipitation over India that will be of great help in management and mitigation of hazards on natural resources and infrastructure.

Loss & Damage (L&D) is the category of policies to address adverse climatic impacts beyond our ability at a given time to adapt to climate change. Policies addressing mitigation, adaptation, and L&D form three pillars of climate policy. While discussions of causal responsibility often group together these three pillars, I distinguish these pillars in a new, time-dependent way and argue that the causal responsibility for L&D is different from the other pillars. The primary reason is that time-dependent causal responsibility for adaptation and mitigation is shared but this is not the case with L&D; by the definition of L&D, there is no possibility of preventative intervening actions between the given time and the impacts. Furthermore, there may be may be cosmopolitan or other climate-independent distributive grounds to contribute to international adaptation meaning that causal responsibility for adaptation may plausibly be shared, unlike with L&D. In contrast, impacts requiring L&D are caused by historical climate emissions alone and would appear to be the responsibility of historical emitters alone. Finally, L&D differs from mitigation in that, for some impacts, historical emissions have led to those impacts becoming beyond adaptive capacity (that is, in the category of L&D) whereas, for all impacts, historical emissions have not led to mitigable impacts becoming non-mitigable. Among other implications of this account, since (a) the distinction between L&D and adaptation depends on the capacities at a given point and (b) the historical emitters on this account have incentive to avoid impacts being classed as L&D (since these are unshared responsibilities), historical emitters have incentive to show that impacts lie within our mitigative or adaptive capacities.

Earthquake stress drop is an important source parameter that directly links to strong ground motion. However, estimating stress drops is often challenging due to many factors, such as data limitations, methodology, and attenuation. Different studies may yield highly inconsistent stress drop values for the same earthquakes, leading to different interpretations of stress drop scaling and spatial patterns. Stress drop is also important for interpretation of earthquake triggering processes. In particular, the roles of foreshocks have continuous debate, and some recent studies show that detailed source parameter analysis is the key. In this study, we combine analyses of three sequences in Southern California using different methods to investigate the resolution of stress drops and the roles of foreshocks. The three sequences include the M7 El-Mayor Cucapah, the 2012 M5 Brawley swarm, and the 2019 M7 Ridgecrest sequence. The Ridgecrest sequence will participate in the Community Stress Drop Validation Study. For each sequence, we apply an improved stacking method and a spectral ratio method. We will use different types of waves: P-wave, S-wave, and Coda-wave. Stress drop results from this study will be compared with available previous studies. We will first discuss the influence of wave type and methodology on stress drop estimations, then we will investigate detailed stress interactions between foreshocks and the mainshock for different types of earthquake sequences (i.e., mainshock-aftershock and swarms that involve aseismic slip or fluid).