Each tab represents one panel from this work with a description of what the data represents. (XLSX)

Dataset

This dataset was created by Amedeo Ercole

Released under CC0: Public Domain

Contents

Data represents key inputs and modeled outputs of the published journal article. Citation information for this dataset can be found in Data.gov's References section.

This dataset contains the well-location information and water-level measurements for 930 wells from 10 confined aquifers of the New Jersey Coastal Plain. Well-location and water-level data is publicly available from the U.S. Geological Survey (USGS) Water Data for the Nation database (USGS, 2025). Water-level measurements from 3 wells in Pennsylvania and reported water-level measurements from 18 wells in Delaware (Delaware Geological Survey [variously dated]) also are included. All measurements were collected from mid-October 2018 to early April 2019 and used to construct potentiometric-surface maps for the confined Cohansey aquifer, Rio Grande water-bearing zone, Atlantic City 800-foot sand, Piney Point aquifer, Vincentown aquifer, Wenonah-Mount Laurel aquifer, Englishtown aquifer system, and upper, middle, and lower aquifers of the Potomac-Raritan-Magothy (PRM) aquifer system. This dataset also contains shapefiles of the potentiometric surface contours for the 10 confined aquife ...

A key challenge in neuroimaging remains to understand where, when and now particularly how human brain networks compute over sensory inputs to achieve behavior. To study such dynamic algorithms from mass neural signals, we recorded the magnetoencephalographic (MEG) activity of participants who resolved the classic XOR, OR and AND functions as overt behavioral tasks (N = 10 participants/task, N-of-1 replications). Each function requires a different computation over the same inputs to produce the task- specific behavioral outputs. In each task, we found that source-localized MEG activity progresses through four computational stages identified within individual participants: (1) initial contra-lateral representation of each visual input in occipital cortex, (2) a joint linearly combined representation of both inputs in midline occipital cortex and right fusiform gyrus, followed by (3) nonlinear task-dependent input integration in temporal-parietal cortex and finally (4) behavioral response...

Data representing crisis contacts made by officers of the Seattle Police Department. Data is denormalized to represent the one to many relationship between the record and the reported disposition of the contact.

USE CAUTION WHEN COUNTING

Graph data represents complex relationships across diverse domains, from social networks to healthcare and chemical sciences. However, real-world graph data often spans multiple modalities, including time-varying signals from sensors, semantic information from textual representations, and domain-specific encodings. This dissertation introduces innovative multimodal learning techniques for graph-based predictive modeling, addressing the intricate nature of these multidimensional data representations. The research systematically advances graph learning through innovative methodological approaches across three critical modalities. Initially, we establish robust graph-based methodological foundations through advanced techniques including prompt tuning for heterogeneous graphs and a comprehensive framework for imbalanced learning on graph data. we then extend these methods to time series analysis, demonstrating their practical utility through applications such as hierarchical spatio-temporal modeling for COVID-19 forecasting and graph-based density estimation for anomaly detection in unmanned aerial systems. Finally, we explore textual representations of graphs in the chemical domain, reformulating reaction yield prediction as an imbalanced regression problem to enhance performance in underrepresented high-yield regions critical to chemists.

This dataset represents land cover mapping, physical habitat measurements, continuous hydrology measurements, salt tracer measurements, and benthic macroinvertebrate sample scores from nine small streams in unincorporated King County within the Puget Sound region of Washington State. These data were collected during two periods, 2008 – 2012/2013 and 2018 – 2022, as part of a study to evaluate the performance of King County’s land use regulations at protecting stream ecosystems. Six of the streams drained watersheds that were developed or developable and were subject to King County’s land use regulations. Three of the streams drained watersheds that were largely protected from development and served as references for comparison. The initial data collection (2008 – 2012/2013) is described in a report titled, “Assessing Land Use Effects and Regulatory Effectiveness on Streams in Rural Watersheds of King County, Washington,” published in 2014. An analysis of the two combined datasets is described in a report titled, “The Effects of the Land Use Regulatory Framework on Stream Ecosystems in Unincorporated King County Watersheds,” published in 2025. See these reports for details about the sampling methods, study results, and what these data represent. Below we briefly describe the types of data included in this dataset. For questions about these data, please contact James Bower (james.bower@kingcounty.gov), Aaron David (adavid@kingcounty.gov), Ian Higgins (ihiggins@kingcounty.gov), or Rebekah Stiling (rstiling@kingcounty.gov). All data were collected by the King County Water and Land Resources Division, Science and Technical Support Section. Land cover mapping of the nine study watersheds was conducted once at the beginning and end of the first period (2007 and 2012) and once at the beginning and end of the second period (2017 and 2022). The land cover data are represented by ‘Land_cover.csv’. Physical habitat measurements were collected once a year within a defined and consistent section of each stream. Physical habitat measurements are represented by ‘Pools.csv’, ‘Reach_lengths.csv’, ‘Substrate.csv’, ‘Thalweg_depths.csv’, and ‘Wood.csv’. Continuous hydrology measurements of stream discharge, water temperature, and conductivity were collected in each stream throughout most years of the study. Continuous hydrology measurements were summarized into daily values and are represented by ‘Hydrology_daily.csv’. Samples of the benthic macroinvertebrate community were collected in each stream during late summer or early fall across all study years. These samples were used to calculate Puget Sound lowlands Benthic-Index of Biotic Integrity scores for each stream and year. Benthic macroinvertebrate sample scores are represented by ‘BIBI.csv’. Salt tracer measurements were conducted in each stream across multiple flows within each year. Salt tracer measurements are represented by ‘Tracer_measurements.csv’. The ‘Variable_names.csv’ file contains a list of each of the variable/field names within each data file, the variable type for each field, and a brief description of what each variable/field represents.

Context

Google added the tfp.sts module un de Probability package and we wanted to test it so we gathered day by day temperature and rain in Donostia / San Sebastian as well as the demand of bikes in the public bike service.

Acknowledgements

Inspiration

This dataset includes synthetically produced data from 10 different cities (Istanbul, Nablus, Chattogram, Cox’s Bazaar, Nairobi, Nakuru, Quito, Kokhana, Rapti and Darussalam) for a future urban context. The data includes physical elements in a city such as buildings, roads, and power networks, as well as social elements such as households and individuals. The dataset contains a maximum of 9 different data types, described below. For some cities power and road network data were not considered due to context specific priorities. landuse: The land use plan data depicting how the land will be zoned and used in the next fifty years within the area or interest. The attributes include the land use type, areal coverage in hectares, maximum population density and existing population. building: Data representing the building footprints that will emerge as a result of the future exposure generation procedure. It includes the attributes of the building such as its identifier number, construction type, number of floors, footprint area, occupation type and construction code level. road nodes: Data representing the points where road segments (edges) are connected to each other, including the identifier number for each node. road edges: Data representing the road segments, including the ID numbers of the starting and ending point (node). power nodes: Data representing the points where power lines (edges) are connected to each other, including the identifier number for each node. power edges: Data representing the power segments, including the including the ID numbers of the starting and ending point (node). household: Data that contains social attributes of a household living in a building. The attributes include number of individuals, income level and commonly used facility ID (such as hospital). individual: Data that contains the attributes of the individuals that are a part of a household. The attributes are age, gender, school ID (if relevant), workplace ID (if relevant) and last attained education level. Distribution table: The future projections for each city that identifies the socio-demographic changes and expected physical development in the next 50 years. The data can be used in geospatial platforms. The nomenclature for the data is as follows: “CitynameFutureExposureDataset/Cityname_CommunityCode_DataType”. This dataset was created as case studies for the Tomorrows Cities: Tomorrowville virtual testbed. It is supported by NERC as part of the GCRF Urban Disaster Risk Hub (NE/S009000/1). Full details about this dataset can be found at https://doi.org/10.5285/cdfea06f-d47c-4967-99d4-cc71bddea45d

Output generated by the RAMP engine for use in the Sector-Coupled Euro-Calliope model. The three datasets in this repository are described briefly here and in more detail in the accompanying README files. Each dataset has an hourly temporal resolution spanning the years 2000 - 2018 (inclusive) and a national spatial resolution spanning 26* - 28** countries in Europe. All datasets are dimensionless; only the profile shapes are used in Euro-Calliope.

Cooking energy demand profiles (ramp-cooking-profiles): Profiles of heat energy demand for cooking in buildings in Europe, stochastically generated using the RAMP model [1]. These profiles are used to distribute annual cooking energy demand in the Euro-Calliope workflow. This dataset covers 28 European countries**.

Electric vehicle plug-in profiles (ramp-ev-plugin-profiles): Profiles of the percentage of parked electric vehicles, stochastically generated using the RAMP-Mobility model [2]. These profiles are used in Euro-Calliope to define the maximum number of electric vehicles that could be plugged in and therefore available to be charged at any given time, assuming controlled (or "smart") charging. This dataset covers 26 European countries*.

Electric vehicle energy consumption profiles (ramp-ev-consumption-profiles): Profiles of the electricity consumption of electric vehicles, stochastically generated using the RAMP-Mobility model [2]. These profiles are aggregated in Euro-Calliope to provide a required percentage of total vehicle electricity demand that must be met in each month. This dataset covers 26 European countries*.

• AUT, BEL, CHE, CZE, DEU, DNK, ESP, EST, FIN, FRA, GBR, HRV, HUN, IRL, ITA, LTU, LUX, LVA, NLD, NOR, POL, PRT, ROU, SVK, SVN, SWE

** (*) + BGR, SRB

*** ALB, MKD, GRC, CYP, BIH, MNE, ISL

[1] Lombardi, Francesco, Sergio Balderrama, Sylvain Quoilin, and Emanuela Colombo. 2019. ‘Generating High-Resolution Multi-Energy Load Profiles for Remote Areas with an Open-Source Stochastic Model’. Energy 177 (June): 433–44. https://doi.org/10.1016/j.energy.2019.04.097.

[2] Mangipinto, Andrea, Francesco Lombardi, Francesco Davide Sanvito, Matija Pavičević, Sylvain Quoilin, and Emanuela Colombo. 2022. ‘Impact of Mass-Scale Deployment of Electric Vehicles and Benefits of Smart Charging across All European Countries’. Applied Energy 312 (April): 118676. https://doi.org/10.1016/j.apenergy.2022.118676.

CWHR Predicted Habitat Models represent areas of predicted suitable habitat for each species within its range. These models are built from the following principal inputs: 1) a statewide, best-available vegetation map (FVEG); 2) GIS data representing a species’ range; 3) the CWHR database of habitat suitability values for over 700 terrestrial vertebrate species. Habitat suitability ranks of Low (non-zero values less than 0.34), Medium (0.34-0.66), and High (greater than 0.66) are based on the maximum suitability value across the 3 species life requisites: reproduction, feeding, and cover. Note that previous versions of these Predicted Habitat Models used an average across the 3 life requisites in order to obtain an overall suitability score for each habitat type and stage class. Habitat suitability scores were developed based on habitat patch sizes greater than 40 acres in size and are best interpreted for habitat patches greater than 200 acres in size. The FVEG landcover dataset is an aggregation of multiple statewide landcover and regional vegetation mapping efforts, conducted at different points in time (approximately 1990 up to time of publishing) and at various resolutions, compiled by the California Department of Forestry and Fire Protection (CALFIRE). FVEG uses the most current and consistent data available for each region of the state. Decision rules were developed that controlled which layers were given priority in areas of overlap. Crosswalks were used to attribute the various data sources according to the CWHR habitat-type classification system. Attributing FVEG with CWHR habitat types allows for the extraction of areas with non-zero suitability values for each species within the bounds of its range, creating a series of maps of predicted suitable habitat which are species-specific. Because FVEG is an amalgam of disparate landcover assessment efforts across the state, the predictive power for determining suitable habitat will vary between species, and possibly even regionally for species which are widely distributed. While these maps represent CDFW’s best estimate of the presence of suitable habitat for any given species in the CWHR system, these maps are also limited by several factors: 1) the accuracy and resolution of vegetation maps in a given region; 2) the dynamic nature of the landscape in which fire and other disturbance events alter conditions at a greater frequency than mapping efforts can track; 3) the currency of expert knowledge, particularly as species adapt to changing land and climate conditions and the shifting of other species’ ranges; 4) the frequency of species-specific surveys across a representative sample of a species’ entire range; 5) metapopulation dynamics, which describes the shifting of populations within their environment as result of numerous types of interactions and responses. CWHR GIS data representing predicted suitable habitat should not be used to indicate the presence or absence of a particular species at any specific site. CWHR predicted habitat models are named according to the 4-character alpha-numeric CWHR ID assigned to each species (5 characters in the case of subspecies or other sub-taxa). There is also a “CWHR Revision Tracking Table” containing a record for each species, its CWHR ID, scientific name, common name, and range and habitat model data revision history. CWHR species range models, predicted habitat models, and GIS data of the statewide distribution of all CWHR habitat types, along with the CWHR revision tracking table, are available for download at https://www.wildlife.ca.gov/Data/CWHR.

Spain Exports of dolls representing only human beings to Ukraine was US$155.59 Thousand during 2006, according to the United Nations COMTRADE database on international trade. Spain Exports of dolls representing only human beings to Ukraine - data, historical chart and statistics - was last updated on November of 2025.

This dataset accompanies geologic map publication " Geologic and geophysical maps of the onshore parts of the Santa Maria and Point Conception 30' x 60' quadrangles, California "; U.S. Geological Survey Scientific Investigations Map 3473. Data presented here include the digital geologic map database, paleontological sample locations and descriptions, and point data sets from magnetic and gravity data. The geologic database includes spatial feature classes and non-spatial tables that collectively contain the geologic information presented in the map plate. Fossil sample localities are included as a point feature class in the geologic map database and as two spreadsheet tables of fossil sample localities and fossil checklists. The location of gravity stations in the map area are compiled and augmented by new measurements of the Santa Maria and northern part of the Point Conception 30 x 60 quadrangle, California. This dataset represents an edited version of Langenheim (2013) U.S. Geological Survey Open-File Report 2013-1282 and includes data that were collected after that publication. Point data representing the isostatic gravity field, gridded at a 200-m spacing, are presented for two reduction densities: 2000 kg/m3 and 2670 kg/m3. Point data representing the magnetic field, gridded at a 200-m spacing, are presented for magnetic data that have been filtered to emphasize shallow-depth and medium-depth magnetic sources. Boundaries of density and magnetic sources were derived from application of the maximum horizontal gradient method for the Santa Maria and Point Conception 30 x 60 quadrangle, California. Point data representing the maximum horizontal gradient boundaries in the magnetic field are presented for the total magnetic field and for magnetic data that have been filtered to emphasize shallow-depth and medium-depth magnetic sources. Point data representing the maximum horizontal gradient boundaries in the gravity field are presented for two reduction densities: 2000 kg/m3 and 2670 kg/m3.

Switzerland Exports of dolls representing only human beings to Paraguay was US$10 during 2001, according to the United Nations COMTRADE database on international trade. Switzerland Exports of dolls representing only human beings to Paraguay - data, historical chart and statistics - was last updated on November of 2025.

Spain Imports from Burkina Faso of Dolls Representing Only Human Beings was US$33 during 2006, according to the United Nations COMTRADE database on international trade. Spain Imports from Burkina Faso of Dolls Representing Only Human Beings - data, historical chart and statistics - was last updated on October of 2025.

Italy Exports of dolls representing only human beings to Indonesia was US$18.5 Thousand during 2006, according to the United Nations COMTRADE database on international trade. Italy Exports of dolls representing only human beings to Indonesia - data, historical chart and statistics - was last updated on November of 2025.

This dataset is comprised of three files containing northing, easting, and elevation ("XYZ") information for light detection and ranging (lidar) data representing the beach topography and sonar data representing near-shore topography of Lake Superior at Minnesota Point, near the Duluth entry, Duluth, Minnesota. The point data is the same as that in LAS files that were used to create the digital elevation models (DEMs) of the approximate 2.15 square kilometer surveyed area. Lidar data were collected September 07, 2022 using a boat mounted Velodyne VLP-16 unit and methodology similar to that described by Huizinga and Wagner (2019). Multibeam sonar data were collected September 06-07, 2022 using a Norbit integrated wide band multibeam system compact (iWBMSc) sonar unit and methodology similar to that described by Richards and Huizinga (2018). Single-beam sonar data were collected September 07, 2022 using a Ceescope echosounder and methodology similar to that described by Wilson and ...

The Urban Heat Island (UHI) effect represents the relatively higher temperatures found in urban areas compared to surrounding rural areas owing to higher proportions of impervious surfaces and the release of waste heat from vehicles and heating and cooling systems. Paved surfaces and built structures tend to absorb shortwave radiation from the sun and release long-wave radiation after a lag of a few hours. The Global Urban Heat Island (UHI) Data Set, 2013, estimates the land surface temperature within urban areas in degrees Celsius (average summer daytime maximum and average summer nighttime minimum) as well as the difference between those temperatures and the temperatures in surrounding rural areas, defined as a 10km buffer around the urban extent. Urban extents are from SEDAC�s Global Rural-Urban Mapping Project, Version 1 (GRUMPv1), and land surface temperatures are from SEDAC�s Global Summer Land Surface Temperature (LST) Grids, 2013, which are derived from the Aqua Level-3 Moderate Resolution Imaging Spectroradiometer (MODIS) Version 5 global daytime and nighttime Land Surface Temperature (LST) 8-day composite data (MYD11A2). For most regions, the UHI data set provides the average daytime maximum (1:30 p.m. overpass) and average nighttime minimum (1:30 a.m. overpass) temperatures in urban and rural areas, and the urban-rural temperature differences, derived from LST data representing a 40-day time-span during July-August (Julian days 185-224) in the northern hemisphere and January-February (Julian days 001-040) in the southern hemisphere. LST grid cells with missing values resulting from high cloud cover in tropical regions were filled with daytime maximum and nighttime minimum LST values from April-May 2013 in the northern hemisphere and December 2013-January 2014 in the southern hemisphere, where available. Some data gaps remain in areas where data were insufficient (e.g., Central Africa).

CWHR Predicted Habitat Models represent areas of predicted suitable habitat for each species within its range. These models are built from the following principal inputs: 1) a statewide, best-available vegetation map (FVEG); 2) GIS data representing a species’ range; 3) the CWHR database of habitat suitability values for over 700 terrestrial vertebrate species. Habitat suitability ranks of Low (non-zero values less than 0.34), Medium (0.34-0.66), and High (greater than 0.66) are based on the maximum suitability value across the 3 species life requisites: reproduction, feeding, and cover. Note that previous versions of these Predicted Habitat Models used an average across the 3 life requisites in order to obtain an overall suitability score for each habitat type and stage class. Habitat suitability scores were developed based on habitat patch sizes greater than 40 acres in size and are best interpreted for habitat patches greater than 200 acres in size. The FVEG landcover dataset is an aggregation of multiple statewide landcover and regional vegetation mapping efforts, conducted at different points in time (approximately 1990 up to time of publishing) and at various resolutions, compiled by the California Department of Forestry and Fire Protection (CALFIRE). FVEG uses the most current and consistent data available for each region of the state. Decision rules were developed that controlled which layers were given priority in areas of overlap. Crosswalks were used to attribute the various data sources according to the CWHR habitat-type classification system. Attributing FVEG with CWHR habitat types allows for the extraction of areas with non-zero suitability values for each species within the bounds of its range, creating a series of maps of predicted suitable habitat which are species-specific. Because FVEG is an amalgam of disparate landcover assessment efforts across the state, the predictive power for determining suitable habitat will vary between species, and possibly even regionally for species which are widely distributed. While these maps represent CDFW’s best estimate of the presence of suitable habitat for any given species in the CWHR system, these maps are also limited by several factors: 1) the accuracy and resolution of vegetation maps in a given region; 2) the dynamic nature of the landscape in which fire and other disturbance events alter conditions at a greater frequency than mapping efforts can track; 3) the currency of expert knowledge, particularly as species adapt to changing land and climate conditions and the shifting of other species’ ranges; 4) the frequency of species-specific surveys across a representative sample of a species’ entire range; 5) metapopulation dynamics, which describes the shifting of populations within their environment as result of numerous types of interactions and responses. CWHR GIS data representing predicted suitable habitat should not be used to indicate the presence or absence of a particular species at any specific site. CWHR predicted habitat models are named according to the 4-character alpha-numeric CWHR ID assigned to each species (5 characters in the case of subspecies or other sub-taxa). There is also a “CWHR Revision Tracking Table” containing a record for each species, its CWHR ID, scientific name, common name, and range and habitat model data revision history. CWHR species range models, predicted habitat models, and GIS data of the statewide distribution of all CWHR habitat types, along with the CWHR revision tracking table, are available for download at https://www.wildlife.ca.gov/Data/CWHR.