Harvard Forest

Data Archive

HF420

Land Conservation and Human Demographics by Census Tract in New England 2014-2018

Related Publications

Data

• hf420-01: New England tract demographics and conservation (preview)
• hf420-02: spatial data metadata
• hf420-03: unbuffered census tract boundaries
• hf420-04: 1 km buffered census tract boundaries
• hf420-05: 2 km buffered census tract boundaries
• hf420-06: 10 km buffered census tract boundaries
• hf420-07: 25 km buffered census tract boundaries
• hf420-08: undeveloped and unprotected land in 1990
• hf420-09: undeveloped and unprotected land circa 2016

Overview

• Lead: Katharine Sims, Neenah Estrella-Luna, Jonathan Thompson
• Investigators: Lucy Lee, Margot Lurie
• Contact: Information Manager
• Start date: 2014
• End date: 2018
• Status: completed
• Location: New England
• Latitude: +40.980144 to +47.459686 degrees
• Longitude: -73.727775 to -66.949895 degrees
• Elevation: 0 to 1917 meter
• Datum: WGS84
• Taxa:
• Release date: 2023
• Language: English
• EML file: knb-lter-hfr.420.2
• DOI: digital object identifier
• EDI: data package
• DataONE: data package
• Related links:
• Project Web Map
• Study type: historical
• Research topic: conservation and management; regional studies
• LTER core area: disturbance patterns
• Keywords: census, conservation, demography, humans, land cover, land use
• Abstract:

  This dataset summarizes land protection, conservation prioritization layer scores, and human demographics within New England communities, defined as census tracts. This dataset was created to identify disparities in land protection according to metrics of social marginalization and assess how incorporating environmental justice criteria into land conservation prioritization systems might change conservation priorities.

• Methods:

  The methods described here are a combination of methods written for Sims et al. 2022 and additions relevant from a data user perspective. For complete references and figures, see Sims et al. 2022 publication. We analyze protected land, characteristics of marginalization, and conservation scores for communities in the New England region. Our unit of analysis is the census tract. For analysis of land protection and conservation scores, we use the 1km buffered tracts in order to account for protected land and conservation values near people living on the edges of census tracts.

  All land protection and conservation variables are calculated within census tracts and within 1km, 2km, 10km, and 25km buffers of those tracts. Shapefiles of all tracts used for analysis are provided as separate objects within this dataset. Data table hf420-01-demographics-conservation.csv can be joined to spatial data of 2018 census tract boundaries by unique ID.

  Protected open space data and measures of access

  Data on protected open space in New England comes from the Harvard Forest/ Highstead POS Dataset Version 1.1 referenced below. The POS dataset includes public land as well as conservation easements and private acquisitions of land by conservation NGOs and small local land trusts. It records a year of protection for a majority of sites (which allows us to investigate historical vs. recent protection). In addition to regional, state, and local sources, this version of the POS layer contains data from the Protected Areas Database of the United States (PAD-US) 2.1, which substantially improved the representation of urban open space due to PAD-US 2.1’s inclusion of the Trust for Public Land’s ParkServe data (U.S. Geological Survey Gap Analysis Project 2020).

  Our primary measure of access to open space is the percent of land protected inside or within a 1 km buffer of each census tract. This is intended to represent protected open space within a community or within a feasible walking distance. We use the percent of land protected rather than an absolute amount because census tracts (and buffers) vary considerably in size. Buffers are important to capture access for people near the boundaries of the census tracts, particularly for communities immediately adjacent to park areas. Some parks receive their own census tracts (code range 9800’s; Boston’s Emerald Necklace area is an example) and would not be counted as part of any community without buffers.

  We apply a series of additional buffers (2km, 10km, 25km) to census tracts to understand the distribution of protected open spaces at varying scales, which may proxy for access to open space on foot versus by car or public transportation. All buffers are clipped to New England boundaries. Further analysis of specific travel times is not possible due to a lack of availability of data on entry points to protected open spaces. While such data exists for many urban areas, it is not available comprehensively at a regional scale.

  Nearby protected open space is assessed altogether as well as by different categories according to ownership (public or private) and year of protection (before or after 1990). If no year of protection is known we assume the year of protection is before 1990. Public open spaces in the dataset (~13% of the region’s land area) include state parks, national parks and national forests, as well as municipal parks and conservation land. Nearly all public open spaces in our region allow public access (limited exceptions include military lands / restricted drinking water supply areas), motivating the separate analysis of their distribution as they are particularly important for recreation access. Private open spaces (~12 % of the region’s land area) are conservation lands owned by individuals, mission-oriented organizations, and more than 300 local and regional land trusts. Given traditions in the region of allowing public access to protected open space, the majority of private lands in our region also allow public access, but data on use rules is not comprehensively available (Daigle et al. 2012). Protected lands that do not allow public access still provide many locally important ecosystem services including flood protection, pollination services, biodiversity protection, cooling during extreme weather events, and amenity value. This motivates our focus on access to all types of protected lands in the main analysis (Foster et al. 2017, Sims et al. 2019, Mandle et al. 2020).

  Our focus on permanently protected open space may underestimate current access to nature in rural communities. However, we feel that the focus on permanently protected land is appropriate because of ongoing and increasing loss of rural open space which threatens future access. A recent report highlights that the loss of natural land cover due to development has also been borne disproportionately by low income and high minority communities (Rowland-Shea et al. 2020). Indeed, the COVID-19 pandemic has led to dramatic pressures on home prices and development in many exurban and rural areas that highlight these threats (Kolko et al. 2021).

  The POS layers were rasterized at 10m resolution prior to analysis, a resolution that is small enough to represent smaller and narrow protected open spaces (e.g. riparian corridors or pedestrian walkways) and large enough to feasibly run analyses at a regional scale. Although census tract boundary data comes with land and water area estimates for each tract, when calculating percent of land protected we define land area using the 2016 NLCD for consistency with land area estimates for buffered census tracts.

  Land Protection Variables

  Land protection variables describe the percent of land area protected within a tract. Land protection is considered overall and in groups: publicly owned, privately owned, before 1990, and 1990 or later. Variables related to land protection include pct_prot, pct_public, pct_private, pct_pre90, pct_post90, pct_post90_pub, pct_post90_priv. These are calculated for each of the four buffer distances as well.

  Characteristics of social marginalization

  We use the five-year 2014-2018 American Community Survey (ACS) estimates for census tracts including total housing units, total population, median household income, race, educational attainment, and English proficiency (Manson et al. 2020b, 2020c). These data are the basis for the four demographic variables critical to our analysis: median household income, percent people of color, percent of people 25 or older without at least a 4-year degree, and percent of people 5 or older who do not speak English at home and speak English less than ‘very well’. In our analysis, the percent people of color variable is measured by including all people who identify with categories other than White, non-Hispanic. This includes people who self-identify as Black, Asian, Native American, Native Hawaiian or Pacific Islander, multiracial, and White Hispanic or Latino. Figure S2 shows the distribution of these demographics within the region.

  Demographic Variables

  The 2014-2018 American Community Survey 5-year estimates are the source of demographic data including: median household income, percent people of color, percent of people English-language isolated, and percent of people 25 or older without at least a 4-year degree. People of color is defined as people who self-identify with any identity other than White non-Hispanic. Variables related to social marginalization include med_hhinc18, pct_poc18, pct_nocoll18, pct_langiso18.

  Definitions of urban, exurban and rural

  Total housing units are divided by the census-provided land-area estimate (ALAND) to determine each tract’s housing density. Tracts are classified as urban, exurban, or rural according to housing density following the cutoffs in Radeloff et al. 2005 (Urban: more than 128 housing units/km2, exurban: 16–128 housing units/km2; rural: less than 16 units/km2). We visually inspected these classifications and reclassified one tract from rural to exurban because it was surrounded by neighboring tracts that were majority urban and was far away from any other rural tracts. We excluded tracts from statistical analysis if they had no housing units, fewer than 100 people, or were missing income data (n = 26). The final dataset comprises N= 3344 tracts: Urban N=2131, exurban N=918 and rural tracts N=295. The distribution of urban, exurban and rural tracts, as well as the excluded tracts, are shown in Figure S3.

  Census Tract Variables

  Several variables describe the geographic characteristics of each census tract, including land area, state, and tract type (urban, exurban, rural) based on housing density. Variables related to tract geography include tract_type, state.

  Ecosystem-based prioritization layers

  To analyze conventional conservation priorities, we calculated average scores from three datasets used to aid in conservation prioritization (Table S1) for land that is still available for conservation. We define available land as land that is undeveloped according to the land cover data and unprotected according to the POS data. Tracts receive a score if they have at least 10 acres of available land according to each layer.

  The rasters used to define available land in 1990 and “2016ish” are provided with this dataset as data objects. In the CSV, we use “2016ish” when referring to available land because, while undeveloped conditions are from the 2016 NLCD, the protected areas data (which is used to define unprotected land) was not cut off at 2016. We used the 2016 NLCD because the 2019 update was not available at the time of analysis, and we did not want to exclude recently protected areas from analysis.

  Conservation Variables

  Conservation variables include the average “score” for a conservation prioritization layer within a tract. These scores are calculated on land that is both undeveloped and unprotected (“available” for conservation) so that scores reflect the conservation value of land that could be protected in the future. Conservation layers analyzed relate to long-term resilience, drinking water, and carbon stocks. The number of brownfields within 1km of each tract is also provided as indicative of restoration opportunities.

  Resilience: The Nature Conservancy’s terrestrial resilience layer is a national dataset that ranks the long-term resilience of the land with a relativized, unitless index at 30m resolution (Anderson et al. 2016). We calculate the median and mean index scores within 1km buffered census tracts. Since the layer does not include scores for areas that are classified as developed, we exclude those areas from the average score calculations. The total number of tracts scored is 2,987 out of the 3,344 total.

  Variables related to resilience include mean_resilience, median_resilience_ resilient_sqm, pct_avail_resilient. These are calculated within the four buffer distances as well.

  Clean drinking water: USDA’s Forests to Faucets 2.0 Assessment is a national layer that ranks sub-watershed (HUC12) importance for cleaning drinking water using a unitless index (U.S. Department of Agriculture Forest Service 2018). We use the relative importance which reflects the percentile rank of each watershed at the national level and allows for comparison across watersheds. We rasterize this layer at 30m resolution for analysis with ArcGIS and calculate the mean and median relative importance scores. The total number of tracts scored is 3,212 out of the 3,344 total.

  Variables related to clean drinking water include mean_f2f, median_f2f, f2f_sqm, pct_avail_f2f. These are calculated within the four buffer distances as well.

  Carbon: The National Biomass and Carbon Dataset for the Year 2000 is a collection of national maps including aboveground biomass estimates circa 2000 at 30m resolution (Kellndorfer et al. 2013). We mosaic the maps for zones 65 and 66, which together cover all of New England, and use the mean of each map’s pixels in areas of overlap. We estimate metric tons of carbon per hectare (C Mg/ha) using the map’s original units of kg/m2 x 10. To do so, we first estimate biomass within each 1km buffered census tract by multiplying the sum of pixels on undeveloped and unprotected land within each tract by 0.09 (which accounts for the proportion of a hectare in each 30m pixel) as described in the dataset user guide (Kellndorfer et al. 2013). We divide this estimate by each tract’s available land area to determine biomass Mg/ha. Finally, we divide by two to convert biomass Mg/ha to C Mg/ha (Schlesinger 1991). The total number of tracts scored is 3,211 out of the 3,344 total.

  Variables related to carbon include sum_biomass_kgm2_10, biomass_metric_tonnes, biomass_tonnes_per_ha, carbon_tonnes_per_ha. These are calculated within the four buffer distances as well.

  Undeveloped land: To estimate undeveloped land in 1990 and 2016, we use Landsat-derived land cover classifications. For 1990, land cover information is derived from two sources: the Continuous Change Detection and Classification (CCDC) algorithm (Olofsson et al. 2016) and the National Land Cover Database (NLCD) (Homer et al 2015). CCDC is an annual product covering the majority of New England but excluding northwest Vermont and northeast Maine. Because it was developed specifically for New England, we considered it to be the more accurate data source for 1990 and used it first where available. Where CCDC data was not available, we filled the remainder of the study area with the 1992 NLCD. For 2016, we used the 2016 NLCD, which has more detail on developed land uses compared to CCDC. For our analysis, undeveloped land includes areas like forests, pasture, farmland, wetlands, and barren land. Undeveloped land does not include developed open space. Undeveloped land is combined with unprotected land to produce rasters of land available (undeveloped and unprotected) for conservation. The rasters of available land for 1990 and 2016ish are provided with this dataset as data objects.

  Variables related to undeveloped land include pct_land, pct_undeveloped_2016, pct_available_2016ish, pct_avail_1990, AVAIL_HA, LAND_AC. All variables except LAND_AC are calculated for the four buffer distances as well.

  Brownfields: To assess redevelopment potential, we analyzed overlap between the EPA’s data on the locations of brownfields and the census tracts (U.S. Environmental Protection Agency 2020). We provide a count because the EPA data is in points.

  Variables related to brownfields include brownfields_1km.

  Removal flag

  While variables were calculated for every census tract in New England, not all tracts were used in analysis due to missing demographic data or low population. A total of 26 tracts were excluded from analysis and can be identified through the removal flag field.

  Variables related to the removal flag include REMOVE.

  Percentile ranks

  For each tract used in analysis (i.e., tracts where REMOVE does not equal 1) we calculate the percentile rank for the four variables used to identify potential environmental justice focus areas. Percentile rankings are on a scale from 0-1 and can be used to query the highest and lowest quartiles (as done by Sims et al. 2021) or to understand where tracts fall within regional distributions for those variables.

  Variables related to percentile ranks include perc_poc, perc_income, perc_prot1km, perc_langiso, perc_nocoll. Quartile ranks for each of these variables are derived from percentile ranks.

  Source Datasets and Citations

  Anderson, M. G., Barnett, A., Clark, M., Prince, J., Olivero Sheldon, A., and Vickery, B. (2016). Resilient and connected landscapes for terrestrial conservation. The Nature Conservancy, Eastern Conservation Science, Eastern Regional Office Boston, MA, 1-149. http://www.conservationgateway.org/ConservationPractices/ClimateChange/Pages/Climate-Resilience.aspx

  Harvard Forest and Highstead Foundation. (2021). New England Protected Open Space (Version 1.1.0): Data set. Zenodo. http://doi.org/10.5281/zenodo.4416179.

  Homer, C., Dewitz, J., Yang, L., Jin, S., Danielson, P., Xian, G., . . . Megown, K. (2015). Completion of the 2011 National Land Cover Database for the conterminous United States–representing a decade of land cover change information. Photogrammetric Engineering and Remote Sensing, 81(5), 345-354. https://www.mrlc.gov/data/nlcd-2016-land-cover-conus

  Kellndorfer, J., Walker, W., Kirsch, K., Fiske, G., Bishop, J., LaPoint, L., . . . Westfall, J. (2013). NACP aboveground biomass and carbon baseline data, V. 2 (NBCD 2000), USA, 2000. ORNL DAAC. https://daac.ornl.gov/cgi-bin/dsviewer.pl?ds_id=1161

  Manson, S., Schroeder, J., Van Riper, D., Kugler, T., and Ruggles, S. (2020a). IPUMS National Historical Geographic Information System: Version 15.0. 2018 TIGER/Line Census Tract Boundaries. Minneapolis, MN: IPUMS.

  Manson, S., Schroeder, J., Van Riper, D., Kugler, T., and Ruggles, S. (2020b). IPUMS National Historical Geographic Information System: Version 15.0. 2018 American Community Survey: 5-Year Data [2014-2018, Tracts and Larger Areas]. Minneapolis, MN: IPUMS. NHGIS table codes: AJ6O, AJ6P, AJ6Q, AJ6R, AJ6S, AJ6T, AJ6U, AJ6V, AKC9, AKGI, AKDK

  Manson, S., Schroeder, J., Van Riper, D., Kugler, T., and Ruggles, S. (2020c). IPUMS National Historical Geographic Information System: Version 15.0. 2018 American Community Survey: 5-Year Data [2014-2018, Block Groups and Larger Areas]. Minneapolis, MN: IPUMS. NHGIS table codes: AJ1S

  U.S. Department of Agriculture Forest Service. (2018). Forests to Faucets 2.0 Assessment. Accessed on 23 November 2020: https://www.fs.fed.us/ecosystemservices/FS_Efforts/forests2faucets.shtml.

  U.S. Environmental Protection Agency. (2020). Brownfields [Dataset]. Accessed on 2 December 2020 from the Geospatial Data Download Service: https://www.epa.gov/frs/geospatial-data-download-service.

• Organization: Harvard Forest. 324 North Main Street, Petersham, MA 01366, USA. Phone (978) 724-3302. Fax (978) 724-3595.

• Project: The Harvard Forest Long-Term Ecological Research (LTER) program examines ecological dynamics in the New England region resulting from natural disturbances, environmental change, and human impacts. (ROR).

• Funding: National Science Foundation LTER grants: DEB-8811764, DEB-9411975, DEB-0080592, DEB-0620443, DEB-1237491, DEB-1832210.

• Use: This dataset is released to the public under Creative Commons CC0 1.0 (No Rights Reserved). Please keep the dataset creators informed of any plans to use the dataset. Consultation with the original investigators is strongly encouraged. Publications and data products that make use of the dataset should include proper acknowledgement.

• License: Creative Commons Zero v1.0 Universal (CC0-1.0)

• Citation: Sims K, Estrella-Luna N, Thompson J. 2023. Land Conservation and Human Demographics by Census Tract in New England 2014-2018. Harvard Forest Data Archive: HF420 (v.2). Environmental Data Initiative: https://doi.org/10.6073/pasta/823727f64e32ad7d0be2c5dc69b0e52f.

Detailed Metadata

hf420-01: New England tract demographics and conservation

1. GISJOIN: 2018 census tract unique id
2. state: census tract state
   • 9: CT
   • 23: ME
   • 25: MA
   • 33: NH
   • 44: RI
   • 50: VT
3. tract_type: type of census tract based on housing density
   • urban: more than 128 housing units/km2
   • exurban: 16–128 housing units/km2
   • rural: fewer than 16 housing units/km2
4. REMOVE: flag to exclude tracts from statistical analysis
   • 0: include
   • 1: exclude
   • 2: consider excluding
5. med_hhinc18: median household income over the last 12 months in 2018 inflation-adjusted dollars (unit: dimensionless / missing value: NA)
6. pct_poc18: percent of people who are not White non-Hispanic (unit: dimensionless / missing value: NA)
7. pct_langiso18: percent of people 5+ years old that speak a language other than English at home and speak English 'less than very well' (unit: dimensionless / missing value: NA)
8. pct_nocoll18: percent of people 25+ years old that do not have a bachelors, graduate, or professional degree (unit: dimensionless / missing value: NA)
9. brownfields_1km: number of brownfields within 1km buffer of census tract (unit: dimensionless / missing value: NA)
10. pct_prot: percent of land area protected within census tract (unit: dimensionless / missing value: NA)
11. pct_public: percent of land area protected and owned by public entity within census tract (unit: dimensionless / missing value: NA)
12. pct_private: percent of land protected and owned by private entity within census tract (unit: dimensionless / missing value: NA)
13. pct_pre90: percent of land protected before 1990 within census tract. includes protected areas with unknown year of protection. (unit: dimensionless / missing value: NA)
14. pct_post90: percent of land protected 1990 or later within census tract (unit: dimensionless / missing value: NA)
15. pct_post90_public: percent of land protected 1990 or later and owned by public entity within census tract (unit: dimensionless / missing value: NA)
16. pct_post90_private: percent of land protected 1990 or later and owned by private entity within census tract (unit: dimensionless / missing value: NA)
17. pct_land: percent of census tract area that is land (unit: dimensionless / missing value: NA)
18. pct_undeveloped_2016: percent of census tract land area that is undeveloped (unit: dimensionless / missing value: NA)
19. pct_available_2016ish: percent of census tract land area that is undeveloped and unprotected ca. 2016 (unit: dimensionless / missing value: NA)
20. pct_avail_1990: percent of census tract land area that is undeveloped and unprotected ca. 1990 (unit: dimensionless / missing value: NA)
21. pct_prot_1km: percent of land area protected within 1 km buffer of census tract (unit: dimensionless / missing value: NA)
22. pct_public_1km: percent of land area protected and owned by public entity within 1 km buffer of census tract (unit: dimensionless / missing value: NA)
23. pct_private_1km: percent of land protected and owned by private entity within 1 km buffer of census tract (unit: dimensionless / missing value: NA)
24. pct_pre90_1km: percent of land protected before 1990 within 1 km buffer of census tract. includes protected areas with unknown year of protection. (unit: dimensionless / missing value: NA)
25. pct_post90_1km: percent of land protected 1990 or later within 1 km buffer of census tract (unit: dimensionless / missing value: NA)
26. pct_post90_public_1km: percent of land protected 1990 or later and owned by public entity within 1 km buffer of census tract (unit: dimensionless / missing value: NA)
27. pct_post90_private_1km: percent of land protected 1990 or later and owned by private entity within 1 km buffer of census tract (unit: dimensionless / missing value: NA)
28. pct_land_1km: percent of 1 km buffer of census tract area that is land (unit: dimensionless / missing value: NA)
29. pct_undeveloped_2016_1km: percent of 1 km buffer of census tract land area that is undeveloped (unit: dimensionless / missing value: NA)
30. pct_available_2016ish_1km: percent of 1 km buffer of census tract land area that is undeveloped and unprotected ca. 2016 (unit: dimensionless / missing value: NA)
31. pct_avail_1990_1km: percent of 1 km buffer of census tract land area that is undeveloped and unprotected ca. 1990 (unit: dimensionless / missing value: NA)
32. pct_prot_2km: percent of land area protected within 2 km buffer of census tract (unit: dimensionless / missing value: NA)
33. pct_public_2km: percent of land area protected and owned by public entity within 2 km buffer of census tract (unit: dimensionless / missing value: NA)
34. pct_private_2km: percent of land protected and owned by private entity within 2 km buffer of census tract (unit: dimensionless / missing value: NA)
35. pct_pre90_2km: percent of land protected before 1990 within 2 km buffer of census tract. includes protected areas with unknown year of protection. (unit: dimensionless / missing value: NA)
36. pct_post90_2km: percent of land protected 1990 or later within 2 km buffer of census tract (unit: dimensionless / missing value: NA)
37. pct_post90_public_2km: percent of land protected 1990 or later and owned by public entity within 2 km buffer of census tract (unit: dimensionless / missing value: NA)
38. pct_post90_private_2km: percent of land protected 1990 or later and owned by private entity within 2 km buffer of census tract (unit: dimensionless / missing value: NA)
39. pct_land_2km: percent of 2 km buffer of census tract area that is land (unit: dimensionless / missing value: NA)
40. pct_undeveloped_2016_2km: percent of 2 km buffer of census tract land area that is undeveloped (unit: dimensionless / missing value: NA)
41. pct_available_2016ish_2km: percent of 2 km buffer of census tract land area that is undeveloped and unprotected ca. 2016 (unit: dimensionless / missing value: NA)
42. pct_avail_1990_2km: percent of 2 km buffer of census tract land area that is undeveloped and unprotected ca. 1990 (unit: dimensionless / missing value: NA)
43. pct_prot_10km: percent of land area protected within 10 km buffer of census tract (unit: dimensionless / missing value: NA)
44. pct_public_10km: percent of land area protected and owned by public entity within 10 km buffer of census tract (unit: dimensionless / missing value: NA)
45. pct_private_10km: percent of land protected and owned by private entity within 10 km buffer of census tract (unit: dimensionless / missing value: NA)
46. pct_pre90_10km: percent of land protected before 1990 within 10 km buffer of census tract. includes protected areas with unknown year of protection. (unit: dimensionless / missing value: NA)
47. pct_post90_10km: percent of land protected 1990 or later within 10 km buffer of census tract (unit: dimensionless / missing value: NA)
48. pct_post90_public_10km: percent of land protected 1990 or later and owned by public entity within 10 km buffer of census tract (unit: dimensionless / missing value: NA)
49. pct_post90_private_10km: percent of land protected 1990 or later and owned by private entity within 10 km buffer of census tract (unit: dimensionless / missing value: NA)
50. pct_land_10km: percent of 10 km buffer of census tract area that is land (unit: dimensionless / missing value: NA)
51. pct_undeveloped_2016_10km: percent of 10 km buffer of census tract land area that is undeveloped (unit: dimensionless / missing value: NA)
52. pct_available_2016ish_10km: percent of 10 km buffer of census tract land area that is undeveloped and unprotected ca. 2016 (unit: dimensionless / missing value: NA)
53. pct_avail_1990_10km: percent of 10 km buffer of census tract land area that is undeveloped and unprotected ca. 1990 (unit: dimensionless / missing value: NA)
54. pct_prot_25km: percent of land area protected within 25 km buffer of census tract (unit: dimensionless / missing value: NA)
55. pct_public_25km: percent of land area protected and owned by public entity within 25 km buffer of census tract (unit: dimensionless / missing value: NA)
56. pct_private_25km: percent of land protected and owned by private entity within 25 km buffer of census tract (unit: dimensionless / missing value: NA)
57. pct_pre90_25km: percent of land protected before 1990 within 25 km buffer of census tract. includes protected areas with unknown year of protection. (unit: dimensionless / missing value: NA)
58. pct_post90_25km: percent of land protected 1990 or later within 25 km buffer of census tract (unit: dimensionless / missing value: NA)
59. pct_post90_public_25km: percent of land protected 1990 or later and owned by public entity within 25 km buffer of census tract (unit: dimensionless / missing value: NA)
60. pct_post90_private_25km: percent of land protected 1990 or later and owned by private entity within 25 km buffer of census tract (unit: dimensionless / missing value: NA)
61. pct_land_25km: percent of 25 km buffer of census tract area that is land (unit: dimensionless / missing value: NA)
62. pct_undeveloped_2016_25km: percent of 25 km buffer of census tract land area that is undeveloped (unit: dimensionless / missing value: NA)
63. pct_available_2016ish_25km: percent of 25 km buffer of census tract land area that is undeveloped and unprotected ca. 2016 (unit: dimensionless / missing value: NA)
64. pct_avail_1990_25km: percent of 25 km buffer of census tract land area that is undeveloped and unprotected ca. 1990 (unit: dimensionless / missing value: NA)
65. mean_resilience: mean resilience score of undeveloped and unprotected land within census tract (unit: dimensionless / missing value: NA)
66. median_resilience: median resilience score of undeveloped and unprotected land within census tract (unit: dimensionless / missing value: NA)
67. resilient_sqm: area of resilient land used in average resilient score calculations (unit: meterSquared / missing value: NA)
68. pct_avail_resilient: percent of census tract available land used in resilient score averages (unit: dimensionless / missing value: NA)
69. mean_resilience_1km: mean resilience score of undeveloped and unprotected land within 1 km buffer of census tract (unit: dimensionless / missing value: NA)
70. median_resilience_1km: median resilience score of undeveloped and unprotected land within 1 km buffer of census tract (unit: dimensionless / missing value: NA)
71. resilient_sqm_1km: area of resilient land used in average resilient score calculations (unit: meterSquared / missing value: NA)
72. pct_avail_resilient_1km: percent of 1 km buffer of census tract available land used in resilient score averages (unit: dimensionless / missing value: NA)
73. mean_resilience_2km: mean resilience score of undeveloped and unprotected land within 2 km buffer of census tract (unit: dimensionless / missing value: NA)
74. median_resilience_2km: median resilience score of undeveloped and unprotected land within 2 km buffer of census tract (unit: dimensionless / missing value: NA)
75. resilient_sqm_2km: area of resilient land used in average resilient score calculations (unit: meterSquared / missing value: NA)
76. pct_avail_resilient_2km: percent of 2 km buffer of census tract available land used in resilient score averages (unit: dimensionless / missing value: NA)
77. mean_resilience_10km: mean resilience score of undeveloped and unprotected land within 10 km buffer of census tract (unit: dimensionless / missing value: NA)
78. median_resilience_10km: median resilience score of undeveloped and unprotected land within 10 km buffer of census tract (unit: dimensionless / missing value: NA)
79. resilient_sqm_10km: area of resilient land used in average resilient score calculations (unit: meterSquared / missing value: NA)
80. pct_avail_resilient_10km: percent of 10 km buffer of census tract available land used in resilient score averages (unit: dimensionless / missing value: NA)
81. mean_resilience_25km: mean resilience score of undeveloped and unprotected land within 25 km buffer of census tract (unit: dimensionless / missing value: NA)
82. median_resilience_25km: median resilience score of undeveloped and unprotected land within 25 km buffer of census tract (unit: dimensionless / missing value: NA)
83. resilient_sqm_25km: area of resilient land used in average resilient score calculations (unit: meterSquared / missing value: NA)
84. pct_avail_resilient_25km: percent of 25 km buffer of census tract available land used in resilient score averages (unit: dimensionless / missing value: NA)
85. mean_f2f: mean Forests to Faucets watershed relative importance score within census tracts (unit: dimensionless / missing value: NA)
86. median_f2f: median Forests to Faucets watershed relative importance score within census tracts (unit: dimensionless / missing value: NA)
87. f2f_sqm: area of Forests to Faucets watershed data used in mean and median calculations (unit: meterSquared / missing value: NA)
88. pct_avail_f2f: percent of census tract available land used in watershed importance score averages (unit: dimensionless / missing value: NA)
89. mean_f2f_1km: mean Forests to Faucets watershed relative importance score within 1 km buffer of census tracts (unit: dimensionless / missing value: NA)
90. median_f2f_1km: median Forests to Faucets watershed relative importance score within 1 km buffer of census tracts (unit: dimensionless / missing value: NA)
91. f2f_sqm_1km: area of Forests to Faucets watershed data used in mean and median calculations (unit: meterSquared / missing value: NA)
92. pct_avail_f2f_1km: percent of 1 km buffer of census tract available land used in watershed importance score averages (unit: dimensionless / missing value: NA)
93. mean_f2f_2km: mean Forests to Faucets watershed relative importance score within 2 km buffer of census tracts (unit: dimensionless / missing value: NA)
94. median_f2f_2km: median Forests to Faucets watershed relative importance score within 2 km buffer of census tracts (unit: dimensionless / missing value: NA)
95. f2f_sqm_2km: area of Forests to Faucets watershed data used in mean and median calculations (unit: meterSquared / missing value: NA)
96. pct_avail_f2f_2km: percent of 2 km buffer of census tract available land used in watershed importance score averages (unit: dimensionless / missing value: NA)
97. mean_f2f_10km: mean Forests to Faucets watershed relative importance score within 10 km buffer of census tracts (unit: dimensionless / missing value: NA)
98. median_f2f_10km: median Forests to Faucets watershed relative importance score within 10 km buffer of census tracts (unit: dimensionless / missing value: NA)
99. f2f_sqm_10km: area of Forests to Faucets watershed data used in mean and median calculations (unit: meterSquared / missing value: NA)
100. pct_avail_f2f_10km: percent of 10 km buffer of census tract available land used in watershed importance score averages (unit: dimensionless / missing value: NA)
101. mean_f2f_25km: mean Forests to Faucets watershed relative importance score within 25 km buffer of census tracts (unit: dimensionless / missing value: NA)
102. median_f2f_25km: median Forests to Faucets watershed relative importance score within 25 km buffer of census tracts (unit: dimensionless / missing value: NA)
103. f2f_sqm_25km: area of Forests to Faucets watershed data used in mean and median calculations (unit: meterSquared / missing value: NA)
104. pct_avail_f2f_25km: percent of 25 km buffer of census tract available land used in watershed importance score averages (unit: dimensionless / missing value: NA)
105. sum_biomass_kgm2_10: total biomass of available land in kg/m2 times ten within census tract (unit: kilogramPerMeterSquared / missing value: NA)
106. biomass_metric_tonnes: metric tonnes of biomass on undeveloped and unprotected land in the census tract (unit: tonne / missing value: NA)
107. biomass_tonnes_per_ha: metric tonnes of biomass per hectare of undeveloped and unprotected land in the census tract (unit: tonnePerHectare / missing value: NA)
108. carbon_tonnes_per_ha: metric tonnes of carbon per hectare of undeveloped and unprotected land in the census tract (unit: tonnePerHectare / missing value: NA)
109. carbon_sqm: area of carbon used in calculations (unit: meterSquared / missing value: NA)
110. pct_avail_carbon: percent of census tract available land used in carbon estimation (unit: dimensionless / missing value: NA)
111. sum_biomass_kgm2_10_1km: total biomass of available land in kg/m2 times ten within 1 km buffer of census tract (unit: kilogramPerMeterSquared / missing value: NA)
112. biomass_metric_tonnes_1km: metric tonnes of biomass on undeveloped and unprotected land in the 1 km buffer of census tract (unit: tonne / missing value: NA)
113. biomass_tonnes_per_ha_1km: metric tonnes of biomass per hectare of undeveloped and unprotected land in the 1 km buffer of census tract (unit: tonnePerHectare / missing value: NA)
114. carbon_tonnes_per_ha_1km: metric tonnes of carbon per hectare of undeveloped and unprotected land in the 1 km buffer of census tract (unit: tonnePerHectare / missing value: NA)
115. carbon_sqm_1km: area of carbon used in calculations (unit: meterSquared / missing value: NA)
116. pct_avail_carbon_1km: percent of 1 km buffer of census tract available land used in carbon estimation (unit: dimensionless / missing value: NA)
117. sum_biomass_kgm2_10_2km: total biomass of available land in kg/m2 times ten within 2 km buffer of census tract (unit: kilogramPerMeterSquared / missing value: NA)
118. biomass_metric_tonnes_2km: metric tonnes of biomass on undeveloped and unprotected land in the 2 km buffer of census tract (unit: tonne / missing value: NA)
119. biomass_tonnes_per_ha_2km: metric tonnes of biomass per hectare of undeveloped and unprotected land in the 2 km buffer of census tract (unit: tonnePerHectare / missing value: NA)
120. carbon_tonnes_per_ha_2km: metric tonnes of carbon per hectare of undeveloped and unprotected land in the 2 km buffer of census tract (unit: tonnePerHectare / missing value: NA)
121. carbon_sqm_2km: area of carbon used in calculations (unit: meterSquared / missing value: NA)
122. pct_avail_carbon_2km: percent of 2 km buffer of census tract available land used in carbon estimation (unit: dimensionless / missing value: NA)
123. sum_biomass_kgm2_10_10km: total biomass of available land in kg/m2 times ten within 10 km buffer of census tract (unit: kilogramPerMeterSquared / missing value: NA)
124. biomass_metric_tonnes_10km: metric tonnes of biomass on undeveloped and unprotected land in the 10 km buffer of census tract (unit: tonne / missing value: NA)
125. biomass_tonnes_per_ha_10km: metric tonnes of biomass per hectare of undeveloped and unprotected land in the 10 km buffer of census tract (unit: tonnePerHectare / missing value: NA)
126. carbon_tonnes_per_ha_10km: metric tonnes of carbon per hectare of undeveloped and unprotected land in the 10 km buffer of census tract (unit: tonnePerHectare / missing value: NA)
127. carbon_sqm_10km: area of carbon used in calculations (unit: meterSquared / missing value: NA)
128. pct_avail_carbon_10km: percent of 10 km buffer of census tract available land used in carbon estimation (unit: dimensionless / missing value: NA)
129. sum_biomass_kgm2_10_25km: total biomass of available land in kg/m2 times ten within 25 km buffer of census tract (unit: kilogramPerMeterSquared / missing value: NA)
130. biomass_metric_tonnes_25km: metric tonnes of biomass on undeveloped and unprotected land in the 25 km buffer of census tract (unit: tonne / missing value: NA)
131. biomass_tonnes_per_ha_25km: metric tonnes of biomass per hectare of undeveloped and unprotected land in the 25 km buffer of census tract (unit: tonnePerHectare / missing value: NA)
132. carbon_tonnes_per_ha_25km: metric tonnes of carbon per hectare of undeveloped and unprotected land in the 25 km buffer of census tract (unit: tonnePerHectare / missing value: NA)
133. carbon_sqm_25km: area of carbon used in calculations (unit: meterSquared / missing value: NA)
134. pct_avail_carbon_25km: percent of 25 km buffer of census tract available land used in carbon estimation (unit: dimensionless / missing value: NA)
135. AVAIL_HA: hectares of undeveloped and unprotected land within the census tract (unit: hectare / missing value: NA)
136. AVAIL_HA_1km: hectares of undeveloped and unprotected land within the 1 km buffer of census tract (unit: hectare / missing value: NA)
137. AVAIL_HA_2km: hectares of undeveloped and unprotected land within the 2 km buffer of census tract (unit: hectare / missing value: NA)
138. AVAIL_HA_10km: hectares of undeveloped and unprotected land within the 10 km buffer of census tract (unit: hectare / missing value: NA)
139. AVAIL_HA_25km: hectares of undeveloped and unprotected land within the 25 km buffer of census tract (unit: hectare / missing value: NA)
140. LAND_AC: acres of land within census tract (unit: acre / missing value: NA)
141. LAT: latitude of tract center in Albers Contiguous Equal Area projection (unit: meter / missing value: NA)
142. LONG: longitude of tract center in Albers Contiguous Equal Area projection (unit: meter / missing value: NA)
143. perc_poc: within-state percentile rank for % people of color (pct_poc18) (unit: dimensionless / missing value: NA)
144. perc_income: within-state percentile rank for median household income (med_hhinc18) (unit: dimensionless / missing value: NA)
145. perc_langiso: within-state percentile rank for % language isolated (pct_langiso18) (unit: dimensionless / missing value: NA)
146. perc_nocoll: within-state percentile rank for % without a 4-year degree (pct_nocoll18) (unit: dimensionless / missing value: NA)
147. perc_prot1km: within-state percentile rank for % of land protected within 1km buffer of tract (pct_prot_1km) (unit: dimensionless / missing value: NA)
148. Q_langiso: within-state quartile rank for % English-language isolated (pct_langiso18) (unit: dimensionless / missing value: NA)
149. Q_nocoll: within-state quartile rank for % without a 4-year degree (pct_nocoll18) (unit: dimensionless / missing value: NA)
150. Q_hhinc: within-state quartile rank for median household income (med_hhinc18) (unit: dimensionless / missing value: NA)
151. Q_poc: within-state quartile rank for % people of color (pct_poc18) (unit: dimensionless / missing value: NA)
152. Q_prot1km: within-state quartile rank for % of land protected within 1km buffer of tract (pct_prot_1km) (unit: dimensionless / missing value: NA)

hf420-02: spatial data metadata

• Compression: none
• Format: pdf
• Type: document

hf420-03: unbuffered census tract boundaries

• Compression: zip
• Format: Esri shapefile
• Type: vector GIS

hf420-04: 1 km buffered census tract boundaries

• Compression: zip
• Format: Esri shapefile
• Type: vector GIS

hf420-05: 2 km buffered census tract boundaries

• Compression: zip
• Format: Esri shapefile
• Type: vector GIS

hf420-06: 10 km buffered census tract boundaries

• Compression: zip
• Format: Esri shapefile
• Type: vector GIS

hf420-07: 25 km buffered census tract boundaries

• Compression: zip
• Format: Esri shapefile
• Type: vector GIS

hf420-08: undeveloped and unprotected land in 1990

• Compression: zip
• Format: tiff
• Type: raster GIS

hf420-09: undeveloped and unprotected land circa 2016

• Compression: zip
• Format: tiff
• Type: raster GIS