ALBERT

Description: 〈span〉〈div〉Abstract〈/div〉We study anthropogenic noise sources seen on seismic recordings along the central section of the San Jacinto fault near Anza, southern California. The strongest signals are caused by freight trains passing through the Coachella Valley north of Anza. Train‐induced transients are observed at distances of up to 50 km from the railway, with durations of up to 20 min, and spectra that are peaked between 3 and 5 Hz. Additionally, truck traffic through the Coachella Valley generates a sustained hum with a similar spectral signature as the train transients but with lower amplitude. We also find that wind turbine activity in northern Baja California introduces a seasonal modulation of 1– to 5‐Hz energy across the Anza network. We show that the observed train‐generated transients can be used to constrain shallow attenuation structure at Anza. Using the results from this study as well as available borehole data, we further evaluate the performance of approaches that have been used to detect nonvolcanic tremor at Anza. We conclude that signals previously identified as spontaneous tremor (〈a href="https://pubs.geoscienceworld.org/bssa#rf21"〉Hutchison and Ghosh, 2017〈/a〉) were probably generated by other nontectonic sources such as trains.〈/span〉

Description: 〈span〉〈div〉ABSTRACT〈/div〉This article gives an overview of machine learning (ML) applications in MyShake—a crowdsourcing global smartphone seismic network. Algorithms from classification, regression, and clustering are used in the MyShake system to address various problems, such as artificial neural network (ANN) and convolutional neural network (CNN) to distinguish earthquake motions, spatial–temporal clustering using density‐based spatial clustering of applications with noise (DBSCAN) to detect earthquakes from phone aggregated information, and random forest regression to learn from existing physics‐based relationships. Beyond existing efforts, this article also presents a vision of the role of ML in some new directions and challenges. Using MyShake as an example, this article demonstrates the promising combination of ML and seismology.〈/span〉

Description: Agriculture is a highly dynamic process in space and time, with many applications requiring data with both a relatively high temporal resolution (at least every 8 days) and fine-to-moderate (FTM 〈 100 m) spatial resolution. The relatively infrequent revisit of FTM optical satellite observatories coupled with the impacts of cloud occultation have translated into a barrier for the derivation of agricultural information at the regional-to-global scale. Drawing upon the Group on Earth Observations Global Agricultural Monitoring (GEOGLAM) Initiative’s general satellite Earth observation (EO) requirements for monitoring of major production areas, Whitcraft et al. (this issue) have described where, when, and how frequently satellite data acquisitions are required throughout the agricultural growing season at 0.05°, globally. The majority of areas and times of year require multiple revisits to probabilistically yield a view at least 70%, 80%, 90%, or 95% clear within eight days, something that no present single FTM optical observatory is capable of delivering. As such, there is a great potential to meet these moderate spatial resolution optical data requirements through a multi-space agency/multi-mission constellation approach. This research models the combined revisit capabilities of seven hypothetical constellations made from five satellite sensors—Landsat 7 Enhanced Thematic Mapper (Landsat 7 ETM+), Landsat 8 Operational Land Imager and Thermal Infrared Sensor (Landsat 8 OLI/TIRS), Resourcesat-2 Advanced Wide Field Sensor (Resourcesat-2 AWiFS), Sentinel-2A Multi-Spectral Instrument (MSI), and Sentinel-2B MSI—and compares these capabilities with the revisit frequency requirements for a reasonably cloud-free clear view within eight days throughout the agricultural growing season. Supplementing Landsat 7 and 8 with missions from different space agencies leads to an improved capacity to meet requirements, with Resourcesat-2 providing the largest incremental improvement in requirements met. The best performing constellation can meet 71%–91% of the requirements for a view at least 70% clear, and 45%–68% of requirements for a view at least 95% clear, varying by month. Still, gaps exist in persistently cloudy regions/periods, highlighting the need for data coordination and for consideration of active EO for agricultural monitoring. This research highlights opportunities, but not actual acquisition rates or data availability/access; systematic acquisitions over actively cropped agricultural areas as well as a policy which guarantees continuous access to high quality, interoperable data are essential in the effort to meet EO requirements for agricultural monitoring.

Description: Global agricultural monitoring utilizes a variety of Earth observations (EO) data spanning different spectral, spatial, and temporal resolutions in order to gather information on crop area, type, condition, calendar, and yield, among other applications. Categorical requirements for space-based monitoring of major agricultural production areas have been articulated based on best practices established by the Group on Earth Observation’s (GEO) Global Agricultural Monitoring Community (GEOGLAM) of Practice, in collaboration with the Committee on Earth Observation Satellites (CEOS). We present a method to transform generalized requirements for agricultural monitoring in the context of GEOGLAM into spatially explicit (0.05°) Earth observation (EO) requirements for multiple resolutions of data. This is accomplished through the synthesis of the necessary remote sensing-based datasets concerning where (crop mask, when (growing calendar, and how frequently imagery is required (considering cloud cover impact throughout the agricultural growing season. Beyond this provision of the framework and tools necessary to articulate these requirements, investigated in depth is the requirement for reasonably clear moderate spatial resolution (10–100 m) optical data within 8 days over global within-season croplands of all sizes, a data type prioritized by GEOGLAM and CEOS. Four definitions of “reasonably clear” are investigated: 70%, 80%, 90%, or 95% clear. The revisit frequency required (RFR) for a reasonably clear view varies greatly both geographically and throughout the growing season, as well as with the threshold of acceptable clarity. The global average RFR for a 70% clear view within 8 days is 3.9–4.8 days (depending on the month), 3.0–4.1 days for 80% clear, 2.2–3.3 days for 90% clear, and 1.7–2.6 days for 95% clear. While some areas/times of year require only a single revisit (RFR = 8 days) to meet their reasonably clear requirement, generally the RFR, regardless of clarity threshold, is below to greatly below the 8 day mark, highlighting the need for moderate resolution optical satellite systems or constellations with revisit capabilities more frequent than 8 days. This analysis is providing crucial input for data acquisition planning for agricultural monitoring in the context of GEOGLAM.

Description: This study examines the suitability of 250 m MODIS (MODerate Resolution Imaging Spectroradiometer) data for mapping global cropland extent. A set of 39 multi-year MODIS metrics incorporating four MODIS land bands, NDVI (Normalized Difference Vegetation Index) and thermal data was employed to depict cropland phenology over the study period. Sub-pixel training datasets were used to generate a set of global classification tree models using a bagging methodology, resulting in a global per-pixel cropland probability layer. This product was subsequently thresholded to create a discrete cropland/non-cropland indicator map using data from the USDA-FAS (Foreign Agricultural Service) Production, Supply and Distribution (PSD) database describing per-country acreage of production field crops. Five global land cover products, four of which attempted to map croplands in the context of multiclass land cover classifications, were subsequently used to perform regional evaluations of the global MODIS cropland extent map. The global probability layer was further examined with reference to four principle global food crops: corn, soybeans, wheat and rice. Overall results indicate that the MODIS layer best depicts regions of intensive broadleaf crop production (corn and soybean), both in correspondence with existing maps and in associated high probability matching thresholds. Probability thresholds for wheat-growing regions were lower, while areas of rice production had the lowest associated confidence. Regions absent of agricultural intensification, such as Africa, are poorly characterized regardless of crop type. The results reflect the value of MODIS as a generic global cropland indicator for intensive agriculture production regions, but with little sensitivity in areas of low agricultural intensification. Variability in mapping accuracies between areas dominated by different crop types also points to the desirability of a crop-specific approach rather than attempting to map croplands in aggregate.

Description: 〈span〉〈div〉ABSTRACT〈/div〉MyShake is a growing smartphone‐based network for seismological research applications. We study how dense array analysis of the seismic wavefield recorded by smartphones may enhance microearthquake monitoring in urban environments. In such areas, the microearthquake signal‐to‐noise ratio on smartphones is not well constrained. We address this issue by compiling a seismic noise model for the Los Angeles (LA) metropolitan area using over 500,000 seismograms recorded by stationary phones running MyShake. We confirm that smartphone noise level is reduced during nighttime, and identify strong noise sources such as major traffic highways, the LA airport, and the Long Beach seaport. The noise analysis shows that stationary smartphones are sensitive to human‐induced ground motions, and therefore smartphone‐derived seismograms may be used to infer the elastic properties of the shallow subsurface. We employ array backprojection analysis on synthetic data to estimate what fraction of LA’s smartphone user population is required to install MyShake to enable the location of events whose induced ground motions are below the smartphone noise level. We find that having 0.5% of LA’s population download the MyShake app would be sufficient to accurately locate M〉1 events recorded during nighttime by stationary phones located at epicentral distances 〈5  km. Currently, the MyShake user coverage in LA is approaching a value that will allow us to locate events whose magnitude is near the regional catalog’s magnitude of completeness.〈/span〉

Description: 〈span〉〈div〉ABSTRACT〈/div〉This article gives an overview of machine learning (ML) applications in MyShake—a crowdsourcing global smartphone seismic network. Algorithms from classification, regression, and clustering are used in the MyShake system to address various problems, such as artificial neural network (ANN) and convolutional neural network (CNN) to distinguish earthquake motions, spatial–temporal clustering using density‐based spatial clustering of applications with noise (DBSCAN) to detect earthquakes from phone aggregated information, and random forest regression to learn from existing physics‐based relationships. Beyond existing efforts, this article also presents a vision of the role of ML in some new directions and challenges. Using MyShake as an example, this article demonstrates the promising combination of ML and seismology.〈/span〉

Description: 〈span〉〈div〉ABSTRACT〈/div〉MyShake is a growing smartphone‐based network for seismological research applications. We study how dense array analysis of the seismic wavefield recorded by smartphones may enhance microearthquake monitoring in urban environments. In such areas, the microearthquake signal‐to‐noise ratio on smartphones is not well constrained. We address this issue by compiling a seismic noise model for the Los Angeles (LA) metropolitan area using over 500,000 seismograms recorded by stationary phones running MyShake. We confirm that smartphone noise level is reduced during nighttime, and identify strong noise sources such as major traffic highways, the LA airport, and the Long Beach seaport. The noise analysis shows that stationary smartphones are sensitive to human‐induced ground motions, and therefore smartphone‐derived seismograms may be used to infer the elastic properties of the shallow subsurface. We employ array backprojection analysis on synthetic data to estimate what fraction of LA’s smartphone user population is required to install MyShake to enable the location of events whose induced ground motions are below the smartphone noise level. We find that having 0.5% of LA’s population download the MyShake app would be sufficient to accurately locate M〉1 events recorded during nighttime by stationary phones located at epicentral distances 〈5  km. Currently, the MyShake user coverage in LA is approaching a value that will allow us to locate events whose magnitude is near the regional catalog’s magnitude of completeness.〈/span〉

Description: Policy makers, government planners and agricultural market participants in Pakistan require accurate and timely information about wheat yield and production. Punjab Province is by far the most important wheat producing region in the country. The manual collection of field data and data processing for crop forecasting by the provincial government requires significant amounts of time before official reports can be released. Several studies have shown that wheat yield can be effectively forecast using satellite remote sensing data. In this study, we developed a methodology for estimating wheat yield and area for Punjab Province from freely available Landsat and MODIS satellite imagery approximately six weeks before harvest. Wheat yield was derived by regressing reported yield values against time series of four different peak-season MODIS-derived vegetation indices. We also tested deriving wheat area from the same MODIS time series using a regression-tree approach. Among the four evaluated indices, WDRVI provided more consistent and accurate yield forecasts compared to NDVI, EVI2 and saturation-adjusted normalized difference vegetation index (SANDVI). The lowest RMSE values at the district level for forecast versus reported yield were found when using six or more years of training data. Forecast yield for the 2007/2008 to 2012/2013 growing seasons were within 0.2% and 11.5% of final reported values. Absolute deviations of wheat area and production forecasts from reported values were slightly greater compared to using the previous year's or the three- or six-year moving average values, implying that 250-m MODIS data does not provide sufficient spatial resolution for providing improved wheat area and production forecasts.

Description: 〈span〉〈div〉Abstract〈/div〉We study anthropogenic noise sources seen on seismic recordings along the central section of the San Jacinto fault near Anza, southern California. The strongest signals are caused by freight trains passing through the Coachella Valley north of Anza. Train‐induced transients are observed at distances of up to 50 km from the railway, with durations of up to 20 min, and spectra that are peaked between 3 and 5 Hz. Additionally, truck traffic through the Coachella Valley generates a sustained hum with a similar spectral signature as the train transients but with lower amplitude. We also find that wind turbine activity in northern Baja California introduces a seasonal modulation of 1– to 5‐Hz energy across the Anza network. We show that the observed train‐generated transients can be used to constrain shallow attenuation structure at Anza. Using the results from this study as well as available borehole data, we further evaluate the performance of approaches that have been used to detect nonvolcanic tremor at Anza. We conclude that signals previously identified as spontaneous tremor (〈a href="https://pubs.geoscienceworld.org/bssa#rf21"〉Hutchison and Ghosh, 2017〈/a〉) were probably generated by other nontectonic sources such as trains.〈/span〉