← Run Sets

Non-Motorized Travel Sensitivity

WF-TDM Non-Motorized Enhancement

Bill Hereth

2026-07-02

Why This Matters

  • The WF-TDM has been updated with an enhanced non-motorized (NM) travel component
  • Before using it in planning applications, we need to know: does the model respond to land use changes in a credible way?
  • We ran 13 controlled tests — artificially increasing household and employment density in five test districts — to see how walk and bike trips respond

How We Tested

Test geography: 5 small districts (278 TAZs) across the region

What we changed:

Group Scenarios
Households only 2×, 4×, 12×
Employment only 2×, 4×, 12×
Both HH + EMP 2×, 4×, 12×
2050 land use (local) Base & Centerized
2050 land use (region) Base & Centerized

What we measured: Non-motorized mode share — the share of daily productions walked or biked

Why such extreme multipliers?

The 2×, 4×, and 12× values are intentionally unrealistic. They stress-test the model across a wide range of densities so we can understand its behavior — not to represent actual planning scenarios.

Overall Sensitivity

All trip purposes combined | Modified small districts | Daily productions

Employment Density is the Stronger Driver

At 4× density

Scenario NM Share vs. Base
Base 2019 16.1%
HH only ×4 18.7% +2.6 pct pts
EMP only ×4 31.1% +15.0 pct pts
HH + EMP ×4 29.5% +13.5 pct pts

At 12× density

Scenario NM Share vs. Base
Base 2019 16.1%
HH only ×12 28.5% +12.4 pct pts
EMP only ×12 51.3% +35.2 pct pts
HH + EMP ×12 49.5% +33.4 pct pts

Important

At equal multipliers, employment density produces a bigger non-motorized mode share gain than household density alone — roughly 5–6× bigger at ×4, narrowing to roughly 2.5–3× at ×12 as the response saturates.

Why is HH + EMP lower than EMP alone?

Mode share is a percentage, and household growth adds trips of every kind — including car trips leaving the area — which grows total trip volume faster than it grows walking/biking trips. At ×12, HH + EMP actually has ~4× more non-motorized trips than EMP alone (3.2M vs. 824K), just spread across a much bigger pool of total trips (6.5M vs. 1.6M) — so the percentage comes out slightly lower even though non-motorized activity is far higher.

Sensitivity by Trip Purpose

Home-Based-Other and Non-Home–Based trips are the most sensitive to employment density — these include lunch, errands, and midday travel that become walkable in dense employment environments.

Walk vs. Bike: Almost All Gains are Walking

EMP scenarios: walk gains are 10–100× larger than bike gains.   HH scenarios: both modes increase, but walk still dominates.

How Much Did Land Use Actually Change?

Table 1: Growth from 2019 base to 2050 forecast
Scenario Local HH Local EMP Region HH Region EMP
Base 2019 +0% +0% +0% +0%
Future Today +108% +61% +76% +46%
Future Today - Centerized +118% +61% +76% +46%

Locally, households roughly double (+108%) and employment grows +61% — much more concentrated than the region-wide totals (+76% HH, +46% EMP), since these test districts were picked for their concentrated growth. Centerization redistributes some additional household growth into the districts (+118%) without changing region-wide totals at all.

2050 Forecasts in Context

2050 local forecasts land at ~19.4% — comparable to a 2× multiplier — a moderate, credible response. Region-wide, the gain is smaller: from 11.0% to 11.8% (+0.8 percentage points).

Where Centerization Has an Effect

Districts gaining households (blue) → Non-Motorized share rises.   Districts losing households (orange) → Non-Motorized share falls.

Non-Motorized (NM) Key Takeaways

  1. The model passes the test — NM mode share increases consistently and proportionally with density increases in the expected direction.

  2. Employment density is the primary lever — at equal multipliers, EMP produces 2–2.5× more NM mode share gain than households.

  3. Walk trips, not bike, drive the gains — especially in high-employment scenarios.

  4. Home-Based-Other and Non-Home–Based trips are most sensitive — short-trip purposes respond most to walkable environments.

  5. 2050 forecasts are in a reasonable range — comparable to roughly 2× current density, not extreme.

Summary



View a more detailed summary