There is a block near the shops where the two sidewalks stopped being the same street some time in the last forty years. The north side has six mature trees, planted close enough that their crowns touch, and a strip of grass wide enough to hold them. The south side has parking meters, a utility box, four steel poles, and no trees at all. In August, between roughly eleven and four, the south side is empty. People walk the north side even when their destination is on the other one, then cross at the last moment.
Nobody experiences that as a policy outcome. It feels like a preference, the way choosing the aisle seat feels like a preference. But the trees on the north side were planted by somebody, watered through their first three summers by somebody, and left standing through at least two rounds of sidewalk repair. The absence on the south side is equally deliberate. Both sides had the same budget available to them and only one of them got a canopy.
Two sides of one block in August
Spend twenty minutes at a corner like that in high summer and the behavior sorts itself into patterns. The people who cross to the shaded side earliest are the ones carrying something, minding a child, or using a walking aid. Anyone in a hurry and unburdened takes the direct route and does not notice the choice they skipped.
Shade works on a different scale from most public amenities. A bench holds two or three people, and a crossing signal serves one corner. A single street tree changes the temperature of the six or seven meters around it for anybody who passes, without a form, a fare, or an eligibility rule. That is also why it goes missing so quietly. There is no queue at a tree and no complaint form when one is removed.
Where the map of shade came from
The distribution is rarely random, and in the United States a good deal of it was written down in advance. Jeremy Hoffman, Vivek Shandas, and Nicholas Pendleton compared 1930s federal mortgage security maps with recent summer land surface temperatures across 108 urban areas and published the comparison in the journal Climate in 2020. In 94 percent of the areas they studied, the neighborhoods that had been graded as hazardous for lending were hotter than the ones graded desirable, by about 2.6 degrees Celsius on average and by as much as 7 degrees in the widest cases.
Those grades were drawn nearly a century ago by people who never mentioned trees. What followed was decades of thinner investment, wider roads, fewer planting strips, and a canopy that was never established. Heat is what that arithmetic looks like on a July afternoon. The maps were retired long ago; the temperature pattern kept the appointment.
What 308 studies add up to
The cooling effect itself is modest on paper and large in practice, which is a hard combination to argue in a budget meeting. Reviewing the research, the Environmental Protection Agency reports that one synthesis of 308 studies found urban forests averaging 3.0 degrees Fahrenheit cooler than urban areas without greenery.
Three degrees sounds like a rounding error until you notice what it is measuring. That figure is an average air temperature across whole areas, not the difference between standing in sun and standing in shade, which anybody who has waited at an unshaded crossing can feel in under a minute. The number tells you the block is cooler. It does not capture the person who could finish the walk because of it.
The stop with no roof on it
Transit makes the gap easy to count, because every stop is a fixed place where people are required to stand still. In February 2023 the UCLA Lewis Center for Regional Policy Studies, working with the advocacy group Move LA, published a report titled "Do L.A. Bus Riders Have Shelter from the Elements?" Of the roughly 10,526 Los Angeles Metro stops for which data was available, 26 percent had a shelter. Coverage varied wildly between municipalities in the same county, from 89 percent of stops in the city of Bell to under 10 percent in Beverly Hills, Downey, and Bellflower. Most of those stops sit in places where average summer highs run above 97 degrees Fahrenheit.
Riders in that system are disproportionately people without a car, so the wait is not optional. A schedule that assumes a five-minute wait in the sun is quietly assuming a body that tolerates it. The same logic runs through whether a stop has anywhere to sit.
Who pays for missing shade
The costs land on a predictable list of people, and none of them file reports about it.
- Outdoor workers, whose shift does not move because the forecast did.
- Older walkers, who lose the errands that used to structure a week when the route stops being survivable.
- Children walking to school in the first weeks of term, on a schedule set months earlier.
- Anyone whose commute includes standing, waiting, or transferring in the open.
There is also a maintenance trap specific to trees. A canopy is not a purchase, it is a twenty-year commitment: watering through establishment, pruning clear of the wires, repairing the sidewalk the roots lift, replacing what dies. Shade planted in a grant cycle and abandoned when the grant closes produces a row of dead stakes, which then becomes the local argument against trying again. Lighting has the same failure pattern after dark, which is why a dead lamp changes a route as reliably as a missing tree does.
Walk the sunny side and count
Take a route you already use, at the hottest hour you would normally be out on it, and count four things per block. How many trees are actually casting shade onto the walking surface. How many are dead, stubbed, or missing from an empty planting hole. Whether the bus stop, crossing, or entrance where you have to stand still has any cover. And whether the shaded side is the one people are walking on, which tells you what everybody has already worked out without discussing it.
Then find out who owns the strip between the sidewalk and the curb outside your own door, because in many towns it is public land the adjacent household is expected to maintain, and in others it is the reverse. A dead street tree can be reported the same way a dark lamp can, with a location and a photograph. If you sit on a school board, a transit committee, or a neighborhood association, the useful question is narrow enough to answer: when was the tree inventory last updated, how many stops on our busiest route have cover, and who is watering what we planted last year.