Seattle Dog Park Data

Finding 01Any park vs. a legally usable park.

Trust for Public Land's access metric uses a half-mile walkshed — the distance most people will walk before driving. By that measure, 99% of Seattle residents live within a 10-minute walk of a park. Applying the same metric to Seattle's off-leash areas produces a substantially smaller number.

Finding 02Mapping the OLA gap.

The map shows Seattle's 14 existing off-leash areas with their half-mile walking-distance areas, the two under construction for fall 2026, and four sites in longer-term planning. These walk areas follow Seattle's pedestrian street network and stop at barriers — I-5, the Ship Canal, steep terrain — that a straight-line circle would cross.

Finding 02bWhere citations happen vs. where walkshed reaches.

If off-leash citations concentrate where OLAs aren't, the pattern should be visible in Animal Control's enforcement data. Here is the overlay.

Of 37 uniquely identified parks in the citation data, 26 sit outside any OLA's 10-minute walkshed and 11 sit inside. Most of the inside parks are themselves OLA host parks — Magnuson (367 citations), Woodland (291), Golden Gardens (227), Genesee (152), Westcrest (122) lead — where the citations fall on the non-OLA portions (athletic fields, shorelines, wooded sections).

Two passes, same direction. Park-named only: 71.9% outside (3,089 of 4,299). Combined with 649 street-address citations geocoded via the Census Bureau batch geocoder with Nominatim fallback (464 of 489 unique addresses matched, 94.9%): 73.4% outside (3,631 of 4,948). Reproducible end-to-end via citation_walkshed_analysis.py + geocode_street_addresses.py.

Method: each citation location is tested for whether it falls inside the merged 0.5-mile walk-distance areas (a point-in-polygon test against the network isochrones). Citation source: data/enforcement-citations.csv (PRRs C049204 + C263949). Isochrones: data/walkshed/ola_isochrones.geojson. 114 blank-location rows excluded. Aggregated output: citation-rate-by-walkshed-status.csv.

Finding 02cSeattle's OLAs and TPL's park-priority tiers.

One more check on placement: overlay the 14 OLAs on an independent map of where new parks would help most. Trust for Public Land maintains exactly that map.

What TPL's Park Priority Index is. For every Seattle Census block group, TPL ParkServe publishes a 1–5 composite score combining low-income density, density of people of color, health indicators (CDC PLACES), and heat/environment stress. A block group scoring 5 is where a new park would have the biggest equity and public-health impact; a 1 is already well-served on those dimensions. Seattle spans 606 block groups across the full range. The index is not a measure of existing park coverage and does not mention dogs.

The cross-tab. Sort each block group into four priority bands (1–2, 2–3, 3–4, 4–5), then compute the share of each band that intersects the union of the 14 OLAs' half-mile walksheds.

Finding 02dOLA acreage vs. licensed-dog population, by ZIP.

Findings 01–02c measure geographic access. This one asks whether OLA acreage inside each ZIP is proportional to the number of dogs registered there. If it were, the per-ZIP ratio would be roughly constant.

What the data is. Seattle Animal Shelter publishes every active pet license on Seattle Open Data. Filtered to species = Dog and joined on ZIP, the April 1 2026 snapshot has 26,652 active dog licenses across 138 ZIPs, of which 32 are inside or adjacent to Seattle. Each ZIP's dog count is joined to the OLA acreage physically inside that ZIP (from data/seattle-olas.csv, point-in-polygon against seattleio/seattle-boundaries-data).

Fair-share baseline. If Seattle's 30.7 OLA acres were distributed in proportion to licensed dogs, each ZIP's fair share would be its dog count times the citywide ratio (~54.7 sq ft of OLA per dog, or 1 acre per ~796 dogs). Actual minus fair share gives the deficit (negative) or surplus (positive).

The same data drawn geographically. Darker navy = more licensed dogs in that ZIP; orange polygons are the union of OLA walksheds. The gaps — high-density navy without orange — are the under-served areas.

Two adjacent-but-distinct neighborhoods carry the city's highest dogs-per-OLA-acre ratios.

On neighborhood labels and ZIPs. ZIP codes are the finest geography the Pet Licenses dataset provides, but they don't map cleanly to named neighborhoods. Most Seattle ZIPs span two to five recognizable neighborhoods (98103: Fremont, Wallingford, Green Lake, Phinney Ridge, south Greenwood; 98122: Capitol Hill + Central District). Labels here lead with a primary neighborhood name and cite the ZIP for precision. A few ZIPs (98133, 98155, 98177, 98178, 98146, 98168) extend past Seattle's city line; those are excluded from the deficit chart because OLA acreage is Seattle-only. Licensed-dog counts are floors — compliance estimates put the actual dog population several times higher.

Finding 03How small is small?

The median Seattle OLA is under one acre. The two smallest (Denny at 0.105 ac and Kinnear at 0.124 ac) are roughly the footprint of a basketball court each. Portland's smallest fenced OLA is larger than Seattle's third-smallest.

Finding 03bSpace per resident, space per dog.

Divide Seattle's total parkland by its human population, and divide the OLA footprint by the dog population. The resulting footprints, drawn at the same scale:

What "too small" means, quantified

Three independent authorities publish overlapping standards for the size of a functioning neighborhood dog park and the per-dog capacity a park of a given size can support.

AKC recommendation

≥ 1 acre

Minimum size with 4–6 ft fence, per AKC Establishing a Dog Park in Your Community.

Industry design consensus

1–5 acres

Community dog-park sizing, with 1 acre as the practical floor. Per Parks & Rec Business and municipal design guidelines (Ann Arbor, Fairfax County).

Per-dog capacity

75–100 sq ft

Recommended planning density at peak use. Below this, injury and behavioral-conflict risk rise. Dog Park Size Guide.

Applied to Seattle's 14 OLAs: seven of 14 are below the AKC 1-acre floor. Three are below 0.25 acre. Implied peak capacity at the 100 sq ft per-dog standard, for the smallest sites:

OLA	Area (sq ft)	Cap at 100 sq ft/dog	Cap at 75 sq ft/dog	Below AKC 1-ac min?
Denny Park	4,574	46	61	Yes — 0.105 ac
Kinnear	5,401	54	72	Yes — 0.124 ac
Plymouth Pillars	8,712	87	116	Yes — 0.20 ac
Regrade	13,068	131	174	Yes — 0.30 ac
Magnolia Manor	20,909	209	279	Yes — 0.48 ac
I-5 Colonnade	21,780	218	290	Yes — 0.50 ac
Lower Woodland	32,670	327	436	Yes — 0.75 ac
Golden Gardens	43,560	436	581	At floor — 1.0 ac
Northacres	60,984	610	813	No — 1.4 ac
Blue Dog Pond	74,052	741	987	No — 1.7 ac
Genesee	117,612	1,176	1,568	No — 2.7 ac
Dr. Jose Rizal	174,240	1,742	2,323	No — 4.0 ac
Westcrest	365,904	3,659	4,879	No — 8.4 ac
Magnuson	392,040	3,920	5,227	No — 9.0 ac

Observation

Kinnear's implied peak capacity is 54 dogs at the 100 sq ft standard and 72 dogs at the 75 sq ft floor. SPR signage at the park asks owners with more than 3 dogs to use another OLA. Kinnear is the only designated OLA within a 10-minute walk of most of Queen Anne.

Park signage asking 3+-dog owners to use another OLA. Park entrance, July 2019 — encampment tarp visible (intermittent). Tent against the OLA's back fence (July 2019); intermittent.

×

Sources: AKC · Parks & Rec Business · Ann Arbor Dog Park Guidelines · FCPA Dog Park Standards · Seattle OLA acreage from data/seattle-olas.csv

Finding 04Peer-city acreage, with a methodology note.

Comparing total OLA acreage across cities is harder than comparing counts because cities define "off-leash" differently. Portland has 38 DOLAs (TPL 2025) but most are unfenced voice-control areas. San Francisco counts 42 dog-friendly sites including small Dog Play Areas (DPAs) inside larger parks. Vancouver BC counts 36 including time-restricted areas. Seattle counts only fully-fenced dedicated OLAs.

A best-effort comparison of OLA acreage per capita, with caveats inline:

* Austin's "largest" entry (Walnut Creek Metropolitan Park, 293 ac voice-control natural area) is not a traditional fenced dog park; the unadjusted Austin total of ~680 ac (sometimes cited externally) reduces to ~80 ac on a fenced-comparable basis (the figure used here). Portland's ~85-ac total comes from PP&R's published DOLA inventory and includes unfenced voice-control sites; Portland's largest single fully-fenced site is Gabriel Park's ~5-ac DOLA, well below the totals shown for SF, Vancouver BC, or Austin's largest.
‡ Fort Funston is GGNRA federal land, not SF Rec & Parks property; off-leash access there has been contested under the GGNRA Dog Management Plan. Listed here as the largest off-leash area SF dogs commonly use; SF's largest city-managed dog play area is closer to 4 ac. The per-capita counts still substantially exceed Seattle. Vancouver BC's 168-acre total (from its 2017 People, Parks & Dogs Strategy) reflects time-restricted shared-use areas on beaches and fields.

Finding 05What Seattle's own data says.

In 2016, as part of the research that became the People, Dogs and Parks Strategic Plan, SPR surveyed Seattle dog owners about compliance with leash laws. The results have not been prominent in SPR communications.

Two of five dog owners in the 2016 survey reported illegal off-leashing in parks on a monthly-or-more basis — a self-reported figure, to a survey run by the enforcing agency. SPR's 2023 OLA Expansion Study frames expansion as a goal “to advance [the] goal of creating OLAs that are increasingly within walking distance of residents and allow SPR to fill in geographic gaps where neighborhoods lack access to an OLA.”

Find-It-Fix-It complaints about off-leash dogs in parks have risen year-over-year as Seattle's population has grown and the OLA count has remained flat.

Finding 06Factors associated with illegal off-leash use.

Finding 07Summary of the Part II findings.

The Part I, Part II, and Part III findings stack into a consistent profile:

• Seattle has been at 14 OLAs since 2009. Population grew 34% over the same window.
• Per-capita OLA count is 1.82 per 100,000 — roughly one-third of Portland, San Francisco, and Vancouver BC.
• Per-capita OLA acreage is lower still. Vancouver BC runs ~6.7× Seattle's per-capita acreage on methodology-adjusted numbers.
• 11.7% of Seattle residents live within a 10-minute network walk of an OLA; 99% live within a 10-minute walk of any park.
• Seven of 14 existing OLAs are below one acre; four are below a quarter-acre.
• Several OLAs have documented site-adjacent safety concerns that shape practical use.
• 39% of Seattle dog owners reported illegal off-leash use in parks monthly or more often in SPR's 2016 survey.
• Two new OLAs open fall 2026, bringing the total to 16. This is the first net increase since 2009 — 17 years — and does not close the per-capita gap.
• Three OLAs — Magnuson, Genesee, Westcrest — show working models with visible governance: a site-specific 501(c)(3), system-wide COLA stewardship, and a recent Park District-funded renovation. (Part III, Finding 03a)
• Six peer cities run some form of shared-use, time-zoned, or voice-control access alongside fenced OLAs (Portland, San Francisco, Vancouver BC, New York, Austin, Boise). Seattle's all-fenced, single-designation model is not represented among them. (Peer Cities)
• Non-SPR public lands — Port of Seattle, SDOT right-of-way, SPU watersheds, Seattle Public Schools fields — are largely untapped as potential OLA sites. (Part III)