Field notes: the mile that looked fine in July and failed by March

Every agency has one of these miles. Mine was a stretch of county road climbing out of a creek bottom, two lanes of asphalt laid eight years back, nothing remarkable about it in July. The windshield crew drove it on a clear Tuesday, rated the surface a 3 on the distress scale, wrote "minor transverse cracking, no action needed" in the notes, and moved on to the next forty miles before lunch. By March it had a longitudinal crack running almost the full length of the segment, two alligator-cracked patches wide enough to swallow a tire, and a pothole crew out there twice in six weeks.

Nothing about that July rating was wrong. That's the part that still bothers me. The crew saw what was there to see.

What they couldn't see, doing sixty miles an hour with a clipboard, was what was happening underneath. The base course on that stretch had been taking on moisture for longer than anyone realized, probably through a shoulder that drained wrong after a culvert repair a few years earlier. Dry asphalt in summer heat can sit on top of a saturated base and look completely fine. The surface doesn't tell you about the water underneath it. Freeze-thaw does. Each cycle that winter lifted and dropped that section a little more, and the cracking that had been invisible or nonexistent in July became unmissable by the first spring pothole patrol.

This is the thing about a single annual drive-by: it's a snapshot of whatever the pavement happened to be doing that week, and asphalt doesn't fail on a schedule that cooperates with your survey calendar. A crew rating in July catches the distress that shows up in July. Base failure driven by freeze-thaw shows up in February and March, right when nobody's out there looking, and by the time the summer crew comes back around the segment has usually already gone from "monitor" to "emergency patch."

The fix isn't rating more often, not with a windshield crew anyway. Nobody's budget covers a second full-network pass in March to catch what the freeze-thaw cycle did. What actually helps is treating condition as a network-wide pattern instead of a mile-by-mile judgment call made at driving speed. Early-stage distress, the hairline cracking and the faint textural changes that precede alligatoring, tends to cluster.

Bad drainage design repeats itself at every culvert tie-in on a corridor. A base course that was placed under one contractor shows the same early signs of ravelling block after block, segment after segment, long before any individual mile looks bad enough for a crew to flag it. A windshield survey, rating one mile at a time, has a hard time seeing that clustering. A network-wide pass of VHR imagery is built exactly for that: it reads the whole system at once and shows you where the early distress is concentrated, not just whether one mile crossed a threshold that day.

That's the shift worth making before next budget season. Not abandoning the field crews, who still know things a picture never will, but giving them a network-level picture to go check instead of sending them out to re-rate everything from scratch. Let the imagery flag where cracking, ravelling and patching are piling up across the network, and let the crew spend their time on the segments that picture says deserve a closer look.

That mile out past the creek bottom is resurfaced now. The segment two miles south of it, with the same base course and the same culvert design, is still sitting at a 3. I'd bet on it going the same way this winter, and I'd rather know that in October than find out in March. If you want to see what that kind of year-round picture of your network looks like before the thaw, take a look at what Road Surface Condition can show you.