A Resource for the Protected Areas Database of the United States (PAD-US)

Data Structure & Attributes

Introduction

For GIS users, understanding the technical structure of PAD-US is crucial to effective work. Here are some key elements to understand.

 

Geodatabase Structure

PAD-US is developed as an Esri Geodatabase with six feature classes of boundaries:

  • ‘Fee’ (PADUS2_0Fee)- Lands owned outright by agencies and organizations
  • ‘Easements’ (PADUS2_0Easement) – Lands with conservation or open space easements, held over private lands and public lands (easements sometimes exist over public fee lands) – this data comes directly into PAD-US from the National Conservation Easement Database (NCED)
  • ‘Designations’ (PADUS2_0Designation) – A wide range of areas such as Wilderness Areas, National Monuments, Areas of Critical Environmental Concern, or Wild & Scenic Rivers. These do not represent ownership and often have been defined with different geometry than the ownership they overlay (especially Wilderness Areas). Several designations may overlay one another. ‘Designation’s are usually defined by Federal legislative actions which sometimes include specific maps of boundaries drawn through political processes – often with little or not reference to actual ownership boundaries. In addition, some State managed areas (Wildlife Management Areas, State Parks) may overlap fee owned Federal lands.
  • ‘Marine’ (PADUS2_0Marine) – The National Oceanic and Atmospheric Administration (NOAA) contributes the boundaries of all Marine Protected Areas (MPA) from their MPA Inventory into PAD-US, which also contain State designations. Many Federal and State areas will be represented as Parks or Refuges in the ‘Fee’ feature class, as well as MPAs in the Marine feature class (MPA is another designation applied to the same unit). The Bureau of Ocean Energy Management (BOEM) also provides Federally managed Outer Continental Shelf lands that can be available for natural resource extraction. Many legitimate overlaps exist between MPAs as protection may vary at the surface or substrate, or for different species or seasons. Overlaps also exist between BOEM and NOAA provided data, with MPAs on top in the layering order.
  • ‘Proclamations’ (PADUS2_0Proclamation) – Includes boundaries that provide context for protected areas, such as military lands (DOD). Most are suitable for illustration only as an outline, or exterior extent, as ownership within is not defined. NPS or USFS ‘Proclamation’ boundaries and FWS Approved Acquisition boundaries describe where each agency has authority to acquire lands, as voluntary opportunities arise. These boundaries may be somewhat or even vastly larger than actual agency fee ownership or easement interest. In addition, FWS Approved Acquisition boundaries may include significant limits for acquisition (for example, only 2,000 acres may be acquired within one of the largest mid western boundaries). ‘Proclamation’ and approved acquisition boundaries do NOT portray Federal ownership and often encompass private lands.
  • ‘Combined’ (PADUS2_0Combined) – ‘Fee’, ‘Designation’, and ‘Easement’ data combined into one feature class, and loaded in that order – this is very powerful for analysis but effective use requires great understanding of the overlaps, largely within the ‘Designation’ layer or between ‘Fee’ and ‘Designation’.

Geodatabase illustration

When working with PAD-US Feature Classes, it is important to note that:

  • Feature classes can overlap by design, but also because boundary data are not perfectly aligned to political boundaries or are imprecise (this also leads to slivers).
  • When feature classes overlap by design (for example a Wild & Scenic River corridor running through a Wilderness Area), management policies may not be easy to ascertain and should be confirmed with managing agencies.
  • ‘Proclamation’ areas do not represent ownership. In the case of USFWS, USFS and NPS they define areas where Congress has provided authority to acquire lands (as voluntary opportunities arise). Google and other commercial mapping services often use them to show, for example, National Forest boundaries – this is misleading, as ‘Proclamation’ boundaries can include significant areas of privately-held lands.
  • Any use of PAD-US data for showing recreational lands must be careful to show only those lands that have fee ownership and designations that are appropriate to recreational use – the PAD-US ‘Access’ attribute is a key tool for these applications, defining what is Open, Restricted/Permitted, and Closed for public access. However, as with all PAD-US data, owning and managing agencies will have the most up to date and accurate information on particular sites.

 

Coordinate System

The PAD-US database is maintained in a projected Coordinate System:
USA Contiguous Albers Equal Area Conic USGS version (GCS North American 1983)

 

The horizontal datum requirement is:
Geographic Coordinate System: GCS_North_American_1983

 

Accuracy

PAD-US data is combined from many sources with varying accuracy, but in general should be reliably accurate at 1:24,000 scale. For some States or agencies, accuracy of source data in PAD-US can be as great as 1:5,000-10,000, or even better.

 

Agreement on which ownership data to use to define the boundaries of protected areas is limited. While there is a general national goal of a nationwide parcel layer, we are far from this goal. Even in some States, Statewide parcel boundaries may not fully line up with individual county tax parcel agencies’ data – while others have more seamless, integrated parcel data. The same is true at the Federal level where some agencies may use boundaries based on the Public Land Survey System (PLSS) and others use different frameworks. Resolving these issues is a broader challenge being taken up by a working group in the Federal Geographic Data Committee.

 

In addition to the basic issue of parcel alignment, some State inventories are submitted to PAD-US with Federal lands – however, these lands are removed in the PAD-US publication process and replaced with approved data provided by the individual Federal agencies themselves. This is done to ensure that Federal lands are provided through an authoritative process. But the result can be that for States with coordinated parcel alignments, their data in PAD-US may now have Federal lands that are no longer fully aligned with their local data. Users focused just on those States may instead choose to use that State’s comprehensive dataset, but with the tradeoff that it may or may not have the most current Federal lands data.

 

Finally, while the Federal lands in PAD-US represent authoritative agency data, some Federal agencies do not yet have complete data for all their holdings. These agencies are working on correcting this as resources are available. In a few cases, States may have more complete Federal lands data than Federal agencies – PAD-US users needing high accuracy for such data should coordinate directly State and Federal data sources to resolve such issues.

 

Topology

PAD-US is an aggregated data inventory – it contains scores and scores of data sources which are all integrated into one combined database. The policy of PAD-US is to accept agency data submissions “as is” and simply translate them into the full PAD-US data structure. This means that the boundaries created by any agency or data-steward may not fully align with those of another, creating GIS topology errors (mostly overlaps or slivers).

 

PAD-US has made great progress with version 2.0 in reducing incidental overlaps and slivers among Federal agencies and future editions should continue this trend. Agencies are being encouraged and supported by PAD-US to sort out their ownership overlaps and provide clean source data into PAD-US.

 

 

PAD-US also has a number of feature classes that intentionally overlay one another – for example, some easements overlay fee lands or other easements; many designation or proclamation boundaries overlay fee and/or easement lands, as well as other designations/proclamations. These are not errors – they’re an accurate reflection of the world of protected areas data. But they can create challenges to GIS users. In PAD-US version 2.0, designations and proclamation boundaries are in separate feature classes which has helped address this issue, but the fact remains that overlapping boundaries are a key part of this dataset.

 

Attributes/Data Dictionary

There are approximately 30 core fields in each PAD-US data layer, divided between fields that come from source agencies (‘Local_Owner’, ‘Local_Name’, etc.) and those created by the PAD-US team from domain values (e.g., ‘Designation’ Type) or from defined calculations (e.g., GAP Status Code):

 

 

Many of these fields use domain values (standards for populating these fields) established by PAD-US. You can download a spreadsheet containing all these domain values from the PAD-US Data Manual. Users are strongly encouraged to explore this PAD-US Data Manual to learn more about these fields, domain values, and, for data-stewards, the best practices for creating and attributing them.

 

Common Attribute Questions

Designation Type Questions: 

What is the difference between Local Park and Local Recreation Area?

Generally, Local Parks are parks by name. There is a bit of a gray area with Local Parks and Local Recreation Areas. The general rule is a Local Recreation Area should be a more developed site, specifically for human oriented intensive recreation activities. In that sense, anything “Playground / Tot Lot / Skate Park / Court / Field / Swim Center / etc”, would be a Local Recreation Area because it is a stand-alone area dominated by an intensive recreation activity. Most of these examples would normally not be followed by “Park” in the name, except in some cases (like Skate Park, which is properly assigned to Local Recreation Area). However, areas like Community Parks, City Parks, County Parks, Dog Parks, Marina Parks, etc, should be designated as Local Park because they are not necessarily places that are developed specifically for or dominated by intensive recreation activities.

 

Why are proposed Wilderness Areas included in the standardized Wilderness Study Area designation type?

While proposed or recommended Wilderness is not legally the same thing as a congressionally designated Wilderness Area, they are managed similarly and are grouped together for PAD-US purposes. See ‘Local Designation Type’ for the official designation type provided by managing agencies.

 

Some western states appear to have more federally owned land than expected. Why is this?

Although many western states have high percentages of federally owned land, it may appear to be even higher when viewing the ‘Combined’ feature class layers. This is primarily because grazing allotments and herd management areas were included from Bureau of Land Management (BLM) data (Note: PAD-US does not currently include grazing allotment and herd management area data for the Forest Service). These areas do not represent ownership boundaries, but rather general management boundaries. Grazing allotments for example, while administered by BLM for livestock grazing, often include a mix of federal, state, and/or private land with more detailed allotment management plans. For this reason, grazing allotments and herd management areas are also in the ‘Designation’ rather than the “Fee” Feature Class. Other federal Designations (such as National Monuments) may also include non-federal lands.