data sources

Info about any spatial or attribute data sources

Datasets from GeoData@SciLi: Libraries as Data Creators

While spending February buried under snow here in Providence, I took the opportunity to update several of the data products we create here at GeoData@SciLi. I’ll provide a summary of what we’re working on in this post. The heading for each project links to its GitHub repo, where you can access the datasets and the scripts we wrote for creating them.

My overall vision has always been that library data services should go beyond simply finding public data and purchasing data for students and faculty; we should actively engage in creating value-added products to meet the research and teaching needs of the university. With a dedication to open data, we also contribute to building a data infrastructure that benefits our local communities, and researchers around world. Creating our own projects keeps our technical skills sharp, gives us more in-depth knowledge about working with particular datasets, and exposes us to the practical processing problems our users face, which makes us better at understanding these issues and thus better able to serve them. To ensure that we can maintain and update our datasets, we automate and script as many of our processes as much as possible. The goal is not to build products, but to build processes to create products.

Ocean State Spatial Database

This is our signature product, a geodatabase of basic Rhode Island GIS and tabular data that folks can use as a foundation for building local projects. The idea is to save mappers the trouble of reinventing the wheel every time they want to do state-based research. I’ve honed this idea over a long period of time; as an graduate student at UW twenty years ago I was creating census databases for the Seattle metropolitan area that we published in WAGDA. I expanded this concept at CUNY, where we created and updated the NYC Geodatabase for many years, which included all forms of mass transit data (which wasn’t readily available at the time). For the Rhode Island version, I pivoted to include layers and attributes that would be of interest at a state-level, and was able to re-use many of the scripts and processes I built previously.

The Census TIGER files are the foundation, and we spent time creating suitably generalized base layers from them. Each layer or object is named with a prefix that categorizes and alphabetizes them in a logical order. “a” layers are areal features that represent land areas (counties, cities and towns, tracts, ZCTAs), “b” features are the actual legal boundaries for these areas (not generalized), “c” features are census data tables that can be joined to the a and b features, and “d” features consist of other points and lines (roads, water bodies, schools, hospitals, etc). The database is published in two formats: a Spatialite version for QGIS, and a file geodatabase for ArcGIS.

Ocean State Spatial Database Layers and Sample Map in QGIS — OSSDB Features and Sample Map in QGIS (Hospitals and the Percentage of Business Establishments that are Health Care Services by ZCTA)

Most of the features are fixed to the 2020 census and don’t change. There are two feature sets that we need to update every year. The first set are tables from the American Community Survey (ACS) and ZIP Code Business Pattern (ZBP). We’ve created tables that consist of a selection of variables that would be of broad interest to many users. We use python notebooks to download the data from the Census Bureau’s API. The ACS variable IDs and labels are stored in a spreadsheet that the script reads in, and checks it against the Census Bureau’s variable list for the demographic profile tables, to see if identifiers and labels have changed compared to the previous year. They often change, so the program flags these and we update the spreadsheet to pull the correct variables. We run the program for a specific geography, and the results are stored in a temporary database. For the ZBP data, we crosswalk and aggregate ZIP Codes to ZCTAs to create ZCTA-level data. I have separate scripts for quality control, where we check number of columns, count of rows, and any given variable to data from last year to see if there are any significant differences that could be errors, and another script for copying the data from the temporary database into the new one.

The other set of features we update are points representing schools, colleges and universities, hospitals, and public libraries. The libraries come from a federal source (IMLS PLS survey), while the others come from state sources (schools and colleges from an educational directory, and hospitals from a licensing directory). We use python to access RIDOT’s geocoding API (their parcel or point-based geocoder) to get coordinates for each feature. There’s a lot of exception handling, to deal with bad or non-matching addresses, some of which creep up every year. I store these in a JSON file; the program runs a preliminary check to see if these addresses have been corrected, and if they’re not the program uses the good address stored in the JSON. For quality control, the Detect Dataset Changes tool in the QGIS Processing toolbox allows us to see if features and attributes have changed, and we do extra work to verify the existence of records that have fallen in or out since last year.

Providence Geocoded Crime Incidents

A few years ago I had an excellent undergraduate fellow in the Data Sciences program who created a process for taking all of the police case logs from the Providence Open Data Portal and creating a GIS dataset out of them. We created this dataset for three reasons: the portal contains just the last 180 days and we wanted to create a historic archive, the records did not have coordinates, and the crimes were not standardized. Geocoding was the biggest challenge, as the location information was listed as one of the following: a street intersection, a block number, or a landmark. The script identifies the type of location, and then employs a different procedure for each. Intersections were easy, as we could pass these to the RIDOT geocoder (their street-interpolation geocoder). For block numbers, the program looks at a local file that contains all addresses in the state’s 911 database, which we filter down to just the City of Providence. It finds the matching street, gets the minimum and maximum address numbers within the given block, and computes the centroid between those addresses. For landmarks like Roger Williams Park or Providence Place Mall, we have a local list of major landmarks with coordinates that the program draws from. All non-matching addresses are written to a separate file, and you have the opportunity to add additional landmarks that didn’t match and rerun them. Crimes are matched to the FBI’s uniform categories for violent and non-violent crime, and there’s also an opportunity to update the list if new incident descriptions appear in the data.

We warn users that the matches are not exact, and this needs to be kept in mind when doing any analysis; for every incident we record the match type so users can assess quality. For all of our projects, we provide detailed documentation that explains exactly how the data was created. At this point we have half the data for 2023, and everything for 2024 and 2025. We run the program a few times each year, to ensure that we capture every incident before 180 days elapses.

Providence Census Geography Crosswalk

I wrote about this project when we released it last year; it is a set of relational tables for taking census data published at the tract, block group, and block level, and apportioning and aggregating it to local Providence geographies that include neighborhoods and wards (there’s also a crosswalk for ZCTAs to local geographies, but it’s rather useless as there is little correspondence). We also published a set of reference maps for showing correspondence or lack thereof between the census and local areas.

Map of Census Tracts and Neighborhoods in Providence — Census Tracts (black outlines) and Neighborhoods in Providence

The newest development is that one of my undergraduates used the crosswalk to generate 2020 census demographic profile summaries for neighborhoods and wards, so that users can simply download a pre-compiled set of data without having to do their own crosswalking. Population and household variables were apportioned using total population as a weight, while housing unit variables were apportioned using total housing units. He also generated percent totals for each variable, which required carefully scrutinizing what the proper numerators and denominators should be based on published census results. Python to the rescue again, he used a notebook that read the census tables in from the Ocean State Spatial Database, which saved us the trouble of using the census API. We publish the data tables in the same GitHub repository as the crosswalk.

UN ICSC Retail Price Indexes

I haven’t updated this one yet, but it’s next on the list. I wrote about this project a few years ago; this is a country-level index that documents variation in the cost of living at different UN duty stations. The UN publishes this data at different intervals throughout the year, in macro-driven Excel files that allow you to pull up data for one country at a time. The trick for this project was looping through hundreds of these files, finding the data hidden by the macro, and turning it into a single time series that includes unique identifiers for place, time, and good / service. This project was born from a research request from a PhD student, and we saw the value of building a process to keep it updated and to publish it for others to use. The scripting was done by the first undergraduate student worker I had at Brown, Ethan McIntosh. Thanks to him, I download the new data each year, run the program, and voila, new data!

Conclusion

I hope you found this summary useful, either because you can use these datasets, or you can learn something from one of our scripts and processes that you can apply to your own work. I hope that more academic libraries will embrace the concept of being data creators, and would incorporate this work into their data service models (along with formally contributing to existing initiatives like the Data Rescue Project or the OpenStreetMap). Feel free to reach out with comments and feedback.

HIFLD Open Data Archived in DataLumos

Some good news to end the year: the Data Rescue Project has finished archiving all of the GIS data layers that were in the HIFLD Open portal, which was decommissioned at the end of summer. I wrote a post for the DRP that summarized the work we did, and you can find all the layers in ICPSR’s DataLumos repository, where you can search for and download layers one by one. I also archived the index for the series and a crosswalk that DHS published for locating updated versions of the data from the individual federal agencies that created them. If you wanted to download the entire set in bulk, it can be transferred from the Brown University Library’s GLOBUS endpoint; there are instructions for doing this on our library’s Data Rescue GitHub repo.

This project was an archival one, in that we were taking a final snapshot of what was in the repository before it went offline. In the coming year, I’ll be thinking about approaches for consistently capturing updates, and there are some folks who are interested in creating a community-driven portal to replace the defunct government site. Stay tuned!

2025 has been a tough year. Wishing you all the best for the year to come. – Frank

DataLumos HIFLD Open Archive

Selected Population Profiles: Census Data for Ethnic Groups

I visit courses to guest lecture on census data every semester, and one of the primary topics is immigrant or ethnic communities in the US. There are many different variables in the Census Bureau’s American Community Survey (ACS) that can be used to study different groups: Race, Hispanic or Latino Origin, Ancestry, Place of Birth, and Residency. Each category captures different aspects of identity, and many of these variables are cross-tabulated with others such as citizenship status, education, language, and income. It can be challenging to pull statistics together on ethnic groups, given the different questions the data are drawn from, and the varying degrees of what’s available for one group versus another.

But you learn something new every day. This week, while helping a student I stumbled across summary table S0201, which is the Selected Population Profile table. It is designed to provide summary overviews of specific race, Hispanic origin, ancestry, and place of birth subgroups. It’s published as part of the 1-year ACS, for large geographic areas that have at least 500,000 people (states, metropolitan areas, large counties, big cities), and where the size of the specific population group is at least 65,000. The table includes a broad selection of social, economic, and demographic statistics for each particular group.

We discovered these tables by typing in the name of a group (Cuban, Nigerian, or Polish for example) in the search box for data.census.gov. Table S0201 appeared at the top of the table results, and clicking on it opened the summary table for the group for the entire US for the most recent 1-year dataset (2024 at the time I’m writing this). The name of the group appears in the header row of the table. Clicking on the dataset name and year in the grey box at the top of the table allows you to select previous years.

Dominican Population in the US — Selected Population Profile for Dominicans in the US

Using the Filters on the left, you can narrow the data down to a specific geography and year. You may get no results if either the geographic area or the ethnic or racial group is too small. Besides table S0201, additional detailed tables appear for a few, isolated years (the most recent being 2021).

Dominican Population in NYC — Selected Population Profile for Dominicans in NYC

A more formal approach, which is better for seeing and understanding the full set of possibilities for ethnic groups and their data availability:

At data.census.gov, search for S0201, and select that table. You’ll get the totals for the entire US.
Using the filters on the left, choose Race and Ethnicity – then a racial or ethnic group – then a detailed race or group – then specific categories until you reach a final menu. This gives you the US-wide table for that group (if available).
Alternatively – you could choose Populations and People – Ancestry instead of Race to filter for certain groups. See my explanation below.
Use the filters again to select a specific geographic area (if available) and years.

With either approach, once you have your table, click the More Tools button (…) and download the data. Alternatively, like all of the ACS tables S0201 can be accessed via the Census Bureau’s API.

Filter by Race and Detailed Group — Filter Menu for Race and Ethnicity – Detailed Options

Where does this data come from? It can be generated from several questions on the ACS survey: Hispanic and Race (respectively, with respondents self-identifying a category), Place of Birth (specifically captures first-generation immigrants), and Ancestry (an open ended question about who your ancestors were).

The documentation I found provided just a cursory overview. I discovered additional information that describes recent changes in the race and ancestry questions. Persons identifying as Native American, Asian, or Pacific Islander as a race, or as Hispanic as an ethnicity, have long been able to check or write in a specific ethnic, national, or tribal group (Chinese, Japanese, Cuban, Mexican, Samoan, Apache, etc). People who identified as Black or White did not have this option until the 2020 census, and it looks like the ACS is now catching up with this. This page links to a document that provides an overview of the overlap between race and ancestry in different ACS tables.

The final paragraph in that document describes table S0201, which I’ll quote here in full:

Table S0201 is iterated by both race and ancestry groups. Group names containing the words “alone” or “alone or in any combination” are based on race data, while group names without “alone” or “alone or in any combination” are based on ancestry data. For example, “German alone or in any combination” refers to people who reported one or more responses to the race question such as only German or German and Austrian. “German” (without any further text in the group name) refers to people who reported German in response to the ancestry question.

For example, when I used my first approach and simply searched for Nigerians as a group, the name appeared in the 2024 ACS table simply as Nigerian. This indicates that the data was drawn from the ancestry question. I was also able to flip back to earlier years. But in my second approach, when I searched for the table by its ID number and subsequently chose a racial group, the name appeared as Nigerian (Nigeria) alone, which means the data came from the race table. I couldn’t flip back to earlier periods, as Nigerian wasn’t captured in the race question prior to 2024.

Consider the screenshot below to evaluate the differences. Nigerian alone indicates people who chose just one race (Black) and wrote in Nigerian under their race. Nigerian alone or in any combination indicates any person who wrote Nigerian as a race, could be Black and Nigerian, or Black and White and Nigerian, etc. Finally, Nigerian refers to the ancestry question, where people are asked to identify who their ancestors are, regardless of whether they or their parents have a direct connection to the given place where that group originates.

Comparison of Nigerian categories in S0201 — Comparison of Race alone, Race Alone or in Combination, and Ancestry for Nigerians

Here’s where it gets confusing. If you search for the S0201 table first, and then try filtering by ancestry, the only options that appear are for ethnic or national groups that would traditionally be considered as Black or White within a US context. Places in Europe, Africa, the Middle East, and Central Asia, as well as parts of the world that were initially colonized by these populations (the non-Spanish Caribbean, Australia, Canada, etc). Options for Asians (south, southeast, and east Asia), Pacific Islanders, Native Americans, and any person who identifies as Hispanic or Latino do not appear as ancestry options, as the data for these groups is pulled from elsewhere. So when I tried searching for Chinese, Chinese alone appears in the table, as this data is drawn from the race table. When I searched for Dominican, the term Dominican appears in the table… Hispanic or Latino is not a race, but a separate ethnic category, and Dominican may identify a person of any race who also identifies as Hispanic. This data comes from the Hispanic / Latino origin table.

My interpretation is that data for Table S0201 is drawn from:

The ancestry table (prior to 2024), and either the race or ancestry table (from 2024 forward), for any group that is Black or White within the US context.
The race table for any group that is Asian, Pacific Islander, or Native American (although for smaller groups, ancestry may have been used prior to 2022 or 2023).
The Hispanic / Latino origin table for any group that is of Hispanic ethnicity, regardless of their race.
Place of birth isn’t used for defining groups, but appears as a set of variables within the table so you can identify how many people in the group are first-generation immigrants who were born abroad.

That’s my best guess, based on the available documentation and my interpretation of the estimates as they appear for different groups in this table. I did some traditional web searching, and then also tried asking ChatGPT. After pressing it to answer my question rather than just returning links to the Census Bureau’s standard documentation, it did provide a detailed explanation for the table’s sources. But when I prompted it to provide me with links to documentation from which its explanation was sourced, it froze and did nothing. So much for AI.

Despite this complexity, the Selected Population Profile tables are incredibly useful for obtaining summary statistics for different ethnic groups, and was perfect for the introductory sociology class I visited that was studying immigration and ancestry. Just bear in mind that the availability of S0201 is limited by size of the geographic area as a whole, and the size of the group within that area.

Alternative Sources for US Census Data

Just when we thought the US government couldn’t possibly become more dysfunctional, it shut down completely on Sept 30, 2025. Government websites are not being updated, and many have gone offline. I’ve had trouble accessing data.census.gov; access has been intermittent, and sometimes it has worked with some web browsers but not with others.

In this post I’ll summarize some solid, alternative portals for accessing US census data. I’ve sorted the free resources from the simplest and most basic to the most exhaustive and complex, and I mention a couple of commercial sources at the end. These are websites; the Census Bureau’s API is still working (for now), so if you are using scripts that access its API or use R packages like tidycensus you should still be in business.

Free and Public

Census Reporter
https://censusreporter.org/
Focus: the latest American Community Survey (ACS) data

A non-profit project originally created by journalists, the Census Reporter provides just the most recent ACS data, making it easy to access the latest statistics. Search for a place to get a broad profile with interactive summaries and charts, or search for a topic to download specific tables that include records for all geographies of a particular type, within a specific place. There are also basic mapping capabilities.

Census Reporter Showing ACS Data for Wilmington, DE

Missouri Census Data Center Profiles and Trends
https://mcdc.missouri.edu/
Focus: data from the ACS and decennial profile tables for the entire US

The Census Bureau publishes four profile tables for the ACS and one for the decennial census that are designed to capture a wide selection of variables that are of broad interest to most researchers. The MCDC makes these readily available through a simple interface where you select the time period, summary level, and up to four places to compare in one table, which you can download as a spreadsheet. There are also several handy charts, and separate applications for studying short term trends. Access the apps from the menu on the right-hand side of the page.

Missouri Census Data Center ACS Profiles Showing Data for Providence and Newport, RI

State and Local Government Data Pages
Focus: extracts and applications for that particular area

Hundreds of state, regional, county, and municipal governments create extracts of census data and republish them on their websites, to provide local residents with accessible summaries for their jurisdictions. In most cases these are in spreadsheets or reports, but some places have rich applications, and may recompile census data for geographies of local interest such as neighborhoods. Search for pages for planning agencies, economic development orgs, and open data portals. New York City is a noteworthy example; not only do they provide detailed spreadsheets, they also have the excellent map-based Population FactFinder application. Fairfax County, VA provides spreadsheets, reports, an interactive map, and spreadsheet tools and macros that facilitate working with ACS data.

NYC Population Factfinder Showing ACS Data for Inwood in Northern Manhattan

IPUMS NHGIS
https://www.nhgis.org/
Focus: all contemporary and historic tables and GIS boundary files for the ACS and decennial census

If you need access to everything, this is the place to go. The National Historic Geographic Information System uses an interface similar to the old American Factfinder (or the Advanced Search for data.census.gov). Choose your dataset, survey, year, topic, and geographies, and access all the tables as they were originally published. There is also a limited selection of historical comparison tables (which I’ve written about previously). Given the volume of data, the emphasis is on selecting and downloading the tables; you can see variable definitions, but you can’t preview the statistics. This is also your best option to download GIS boundary files, past and present. You must register to use NHGIS, but accounts are free and the data is available for non-commercial purposes. For users who prefer scripting, there is an API.

IPUMS NHGIS Filtered to Show County Data on Age from the 1990 Census

MCDC Uexplore / Dexter
https://mcdc.missouri.edu/applications/uexplore.html
Focus: create targeted extracts of ACS data and the decennial census back to 1980

Unlike other applications where you download data that’s prepackaged in tables, Uexplore allows you to create targeted, customized extracts where you can pick and choose variables from multiple tables. While the interface looks daunting at first, it’s not bad once you get the hang of it, and it offers tremendous flexibility and ample documentation to get you started. This is a good option for folks who want customized extracts, but are not coders or API users.

Portion of the Filter Interface for MCDC Uexplore / Dexter

Commercial Products

There are some commercial products that are quite good; they add value by bundling data together and utilizing interactive maps for exploration, visualization, and access. The upsides are they are feature rich and easy to use, while the downsides are they hide the fuzziness of ACS estimates by omitting margins of error (making it impossible to gauge reliability), and they require a subscription. Many academic libraries, as well as a few large public ones, do subscribe, so check the list of library databases at your institution to see if they subscribe (the links below take you to the product website, where you can view samples of the applications).

PolicyMap
https://www.policymap.com/
Focus: mapping contemporary census and US government data

PolicyMap bundles 21st century census data, datasets from several government agencies, and a few proprietary series, and lets you easily create thematic maps. You can generate broad reports for areas or custom regions you define, and can download comparison tables by choosing a variable and selecting all geographies within a broader area. It also incorporates some basic analytical GIS functions, and enables you to upload your own coordinate point data.

PolicyMap Displaying ACS Income Data for Providence, RI

Social Explorer
https://www.socialexplorer.com/
Focus: mapping contemporary and historic US census data

Social Explorer allows you to effortlessly create thematic maps of census data from 1790 to the present. You can create a single map, side by side maps for showing comparisons over time, and swipe maps to move back and forth from one period to the other to identify change. You can also compile data for customized regions and generate a variety of reports. There is a separate interface for downloading comparison tables. Beyond the US demographic module are a handful of modules for other datasets (election data for example), as well as census data for other countries, such as Canada and the UK.

Social Explorer Map Displaying ACS Migration Data for Providence, RI

HIFLD Open GIS Portal Shuts Down Aug 26 2025

HIFLD Open, a key repository for accessing US GIS datasets on infrastructure, is shutting down on August 26, 2025. This is a revision from a previous announcement, which said that it would be live until at least Sept 30. The portal provided national layers for schools, power lines, flood plains, and more from one convenient location. DHS provides no sensible explanation for dismantling it, other than saying that hosting the site is no longer a priority for their mission (here’s a copy of an official announcement). In other words, “Public domain data for community preparedness, resiliency, research, and more” is no longer a DHS priority.

The 300 plus datasets in Open HIFLD are largely created and hosted by other agencies, and Open HIFLD was aggregating different feeds into one portal. So, much of the data will still be accessible from the original sources. It will just be harder to find.

DHS has published a crosswalk with links to alternative portals and the source feeds for each dataset, so you can access most of the data once Open HIFLD goes offline. I’ve saved a copy here, in case it also disappears. Most of these sources use ESRI REST APIs. Using ArcGIS Online or Pro, and even QGIS (for example), you can connect to these feeds, get a listing in your contents pane, and drag and drop layers into a project (many of the layers are also available via ArcGIS Online or the Living Atlas if you’re using Arc). Once you’ve added a layer to a project, you can export and save local copies.

QGIS ESRI Rest Services — Adding ArcGIS Rest Server for US Army Corps of Engineers Data in QGIS

If you want to download copies directly from Open HIFLD before it vanishes on Aug 26, I’ve created this spreadsheet with direct links to download pages, and to metadata records when available (some datasets don’t have metadata, and the links will bring you to an empty placeholder). Some datasets have multiple layers, and you’ll need to click on each one in a list to get to it’s download page. In some cases there won’t be a direct download link, and you’ll need to go to the source (a useful exercise, as you’ll need to remember where it is in the future). Alternatively, you can connect to the REST server (before Aug 26, 2025) in QGIS or ArcGIS, drag and drop the layers you want, and then export:

https://services1.arcgis.com/Hp6G80Pky0om7QvQ/ArcGIS/rest/services

I’m coordinating with the Data Rescue Project, and we’re working on downloading copies of everything on Open HIFLD and hosting it elsewhere. I’ll provide an update once this work is complete. Even though most of these datasets will still be available from the original sources, better safe than sorry. There’s no telling what could disappear tomorrow.

The secure HIFLD site for registered users will remain available, but many of the open layers aren’t being migrated there (see the crosswalk for details). The secure site is available to DHS partners, and there are restrictions on who can get an account. It’s not exactly clear what they are, but it seems unlikely that most Open users will be eligible: “These instructions [for accessing a secure account] are for non-DHS users who support a homeland security or homeland defense mission AND whose role requires access to the Geospatial Information Infrastructure (GII) and/or any geospatial dashboards, data, or tools housed on the GII…“

GIS Data Sources Spring Roundup

It’s been awhile since I’ve written a post that showcases different GIS datasets. So in this one, I’ll provide an overview of some free and open data sources that I’ve learned about and worked with this past spring semester. The topics in these series include: global land use and land cover, US heat and temperature, detailed population data for India, and public health in low and middle income countries.

GLAD Global Land Analysis and Discovery

https://glad.umd.edu/dataset

The GLAD lab at the Department of Geographical Sciences at the University of Maryland produces over a dozen GIS datasets related to global land use, land cover, and change in land surface over time. Last semester I had folks who were interested in looking at recent global change in cropland and forest. GLAD publishes rasters that include point-in-time coverage, period averages, and net change and loss over the period 2000 to 2020. Much of the data is generated from LANDSAT, and resolution varies from 30m to 3km. Other series include tropical forest cover and change, tree canopies, forest lost due to fires, a few non-global datasets that focus on specific regions, and LANDSAT imagery that’s been processed so it’s ready for LULC analysis.

Most of the sets have been divided up into tiles and segmented based on what they’re depicting (change in crops, forest, etc). The download process is basic point and click, and for larger sets they provide a list of tifs in a text file so you can automate downloading by writing a basic script. Alternatively, they also publish datasets via Google Earth Engine.

GLAD cropland in Rhode Island — GLAD Cropland Extent in 2019 in QGIS, Zoomed in to Optimal Resolution in SE Rhode Island

US Heat Severity Index

For the past few years, the Trust for Public Land has published an annual heat severity index. This layer represents the relative heat severity for 30m pixels for every city in the United States; depicting where areas of cities are hotter than the average temperature for that same city as a whole (i.e. the surface temperature for each pixel relative to the general air temperature reading for the entire city). Severity is measured on a scale of 1 to 5, with 1 being a relatively mild and 5 being severe heat. The index is generated from a Heat Anomalies raster which they also provide; it contains the relative degrees Fahrenheit difference between any given pixel and the mean heat value for the city in which the pixel is located. Both datasets are generated from 30-meter Landsat 8 imagery, band 10 (ground-level thermal sensor) from summertime images.

The dataset is published as an ArcGIS image service. The easiest way to access it is by to adding it from the Living Atlas to ArcGIS Pro (or Online), and then export the service from there as a raster feature class (while doing so, you can also clip the layer to a smaller area of interest). It’s possible that you can also connect to it as an ArcGIS REST Server in QGIS, but I haven’t tried. While there are files that go back to 2019, the methodology has changed over time, so studying this as a national, annual time series is not appropriate. The coverage area expanded from just large, incorporated cities in earlier years to the entire US in recent years.

US Heat Index in Rhode Island — US Heat Severity Index 2023 in ArcGIS Pro, Providence and Adjacent Areas with Census Blocks

SHRUG – India

https://devdatalab.org/shrug_download/

Created and hosted by the Development Data Lab (a collaborative project created by academic researchers from several universities), the Socioeconomic High-resolution Rural-Urban Geographic Platform for India (SHRUG) is an open access repository consisting of datasets for India’s medium to small geographies (districts, subdistricts, constituencies, towns, and villages), linked together with a set of common geographic IDs. Getting geographically detailed census data for India is challenging as you have to purchase it through 3rd party vendors, and comparing data across time is tough given the complex sets of administrative subdivisions and constant revisions to geographic identifiers. SHRUG makes it easy and open source, providing boundaries from the 2011 census and a unique ID that links geographies together and across time, back to 1991. In addition to the census, there are also environmental and election datasets.

Polygon boundaries can be downloaded as shapefiles or geopackages, and tabular data is available in CSV and DTA (STATA) formats. Researchers can also contribute data created from their own research to the repository.

SHRUG India District Data — SHRUG India Districts Total Population Data from 2011 Census in QGIS

USAID Spatial Data and Population Indicators

USAID published the detailed Demographic and Health Surveys (DHS) as far back as the mid 1980s for many of the world’s low and middle income countries. The surveys captured information about fertility, family planning, maternal and child health, gender, HIV/AIDS, literacy, malaria, nutrition, and sanitation. A selection of different countries were surveyed each year, and for most countries data was captured at two or three different points in time over a 40 year period. While researchers had to submit proposals and request access to the microdata (individual person and household level responses), the agency generated population-level estimates for countries and country subdivisions that were readily downloadable. They also generated rasters that interpolated certain variables across the surface of a country (the header image for this post is a raster of Senegal in 2023, illustrating the percentage of children aged 12-36 months who are vaccinated for eight fundamental diseases, including measles and polio). The rasters, boundary files, and a selection of survey indicators pre-joined to country and subdivision boundaries were published in their Spatial Data Repository. You could access the full range of population indicators as tables from a point and click website, or alternatively via API.

I’m writing in the past tense, as USAID has been decimated and de-funded by DOGE. There is currently no way to request access to the microdata. The summary data is still available on the USAID website (via links in the previous paragraph), but who knows for how long. As part of the Data Rescue Project, I captured both the Spatial Data Repository and the Indicators data, and posted them on DataLumos, an archive of archived federal government datasets. You can download these datasets in bulk from DataLumos, from the links under the title for this section. Unfortunately this series is now an archive of data that will be frozen in time, with no updates expected. The loss of these surveys is not only detrimental to researchers and policymakers, but to millions of the world’s most vulnerable people, whose health and well-being were secured and improved thanks to the information this data provided.

At These Coordinates

Dispatches from the Geospatial Data World