Inspecting trails

Open a trail in the Field App and tap Start inspection. There are five questions, one per screen, and you answer every one with a tap. There is nothing to type, so it works with gloves on and in the rain. Add a photo on any question and it stays with that question. It does not end up in a pile at the end.

It is the same wizard as Inspecting structures, asking about a trail instead of a bridge. One of the questions does something the others do not, and the last section explains it.

The five questions

• Is the trail usable right now? This comes first, because it is the only answer that changes what somebody does in the next ten minutes. If the inspector is interrupted halfway through, you still have the answer that matters.
• How wet is the tread? Dry and dusty, firm and dry, damp but firm, soft or greasy, or puddles and running water.
• Tread: holding up, normal wear, rutting or braiding, or washed out.
• Anything in the way? Clear, blowdown you can get past, or blocked.
• Drains and water bars: running clear, silting up, blocked or ponding, or none on this trail.

Every answer carries its own grade of Good, Marginal or Poor, and the record takes the worst one. A trail with four good answers and a washout is not "mostly fine". It is a trail with a washout. The record says which answer set the grade, so a manager reading the queue knows what to bring before they leave.

Tread and Anything in the way? also offer Doesn't apply, for a section the question is not about. It grades nothing. Passing over a question cannot move the walk's rating either way, and it is never named as the answer that set the grade. It still stays in the record, and that is the point: "we looked, and there is nothing of that kind here" is worth as much as a grade.

The first two questions never offer it. Every trail is either usable or not, and every trail has a surface. The wetness answer is also what reaches the trail-status engine, so a walk that skipped it would tell the engine nothing. Drains and water bars does not offer it either, because None on this trail already says the same thing, and it carries a grade, so it says more.

What happens when you save

A record is kept. It holds who walked the trail, when, every answer and the photos. It is never edited, because a record you can go back and change is not evidence of anything. If you made a mistake, inspect again. That is also what happened on the ground.

The app offers to file work. It never files it for you. If the answers add up to Marginal or Poor, the app offers to Create a maintenance task with the failing answers already written in. You are standing at the problem, which is the easiest moment it will ever be reported, so it takes one tap. The box starts unticked. Under it, the app names any task already open on that trail, so you can see whether somebody has already reported the same thing.

The box lists the findings it will write in, so you can see what the task says before you tick it. A question you passed over is not among them.

Notes (optional) sits above both boxes. Use it for what you saw and what it will take, in your own words. The note belongs to that inspection and shows on that walk's row in the history. It does not change the task or reach anybody's task list.

The trail's own status does not change unless you ask. That is on purpose, and the next section explains why. It is also why an inspection is graded Good, Marginal or Poor instead of Open, Caution or Closed. A walk records what the ground is like. It is not a decision to close a trail, and the different words make it clear which of the two you are doing.

When you want both, the review step offers a second box: Set the trail's status, with Caution and Closed beside it. Like the maintenance box it starts unticked, and for a stronger reason. Filing a task adds work to a list somebody sorts through. Changing a trail's status is a public statement that changes what riders and hikers do today. Tick it and the trail changes as soon as the inspection lands, so it works with no signal like everything else here.

Two things are worth knowing before you rely on it. Only people who can already change trail statuses see it. That is the same permission as the Trail Conditions screen, so volunteers record the walk but do not set the status. And an inspection can set Caution or Closed but can never reopen a trail. Walking one stretch and finding it fine does not prove that whatever closed the trail has been dealt with. Reopening is still a decision somebody makes on the Trail Conditions screen, with the whole picture in front of them.

If your account has weather automation set to open and close trails, the box tells you so. The next weather check writes whatever the forecast says, so a status you set by hand can be replaced within the hour. Switch the setting to close only if you want your decision to stand.

The wetness answer goes to the trail-status engine

This is the part worth knowing.

If you use weather automation, TrailsIQ keeps a running estimate of how much water your ground is holding. It is built from months of daily rainfall, snowmelt and drying. See How to enable weather automation.

An inspection measures the thing that estimate is guessing at. So when somebody reports how wet the tread is, the engine stops estimating for that day and starts from what was seen. It then adds the real rain, snowmelt and drying since. The forecast is not thrown away. It just starts from a better place.

The effect is what you would want. Say the weather points to closing a network, and somebody walked one of its trails yesterday and found it firm and dry. The engine will recommend closing the others and leaving that one open. The recommendation says so in plain words: "Conditions good. Trail wetness 12% from Dana's walk on Aug 30 …"

Only the wetness answer does this. Skip that question and the inspection is still a full record. It simply tells the engine nothing, which is right. An inspection that skipped the question is not a report of average ground.

How long your walk outranks the forecast

A reading does not last forever. It fades out gradually instead of stopping on a set day, and it lasts longer in one direction than the other.

On the day you walk the trail, its conditions are your reading. Each day after, a little more of the forecast comes back. Eventually your walk no longer moves the number and stops being counted:

• A reading wetter than the forecast fades out over about ten days.
• A reading drier than the forecast fades out over about four.

It fades gradually so that conditions never change with no weather behind them. If a walk counted in full yesterday and not at all today, it would look like the app had changed its mind.

The difference between ten days and four is on purpose. Somebody reporting soft or greasy ground has stood in it. That report is specific, and nobody files it by accident. Somebody reporting firm and dry ground is describing the whole trail from wherever they happened to be. That is often the drained ridge, and rarely the low, wet corner that decides the call. Being wrong in the wet direction costs a ride. Being wrong in the dry direction costs tread that takes a season to grow back.

Two things follow that are easy to misread:

• Rain since your walk still counts. A dry reading does not keep a trail open through a storm. The storm is added on top of what you saw, and the trail closes.
• "Wetter" and "drier" are compared with what the forecast would say today, not with how wet the ground felt to you. A dusty reading in August usually agrees with the forecast instead of correcting it, and a reading that agrees is not holding anything up.

Walk the trail again and the new reading counts in full from that day.

Reading the history

In the Field App, open Trails and tap the trail. Its Inspection card shows the latest walk. Tap it to open every walk in its own sheet, newest first, with the rating each one came back with. The question that matters on the trail is whether it is worse than last time, so each walk also shows what changed since the one before. Worse is in red and better is in green. Tap a row to open the answers as they were worded that day, with the photos taken on each one. The history works with no signal from what your phone already holds, and it tells you when you are looking at the stored copy. A walk still waiting to upload appears in the list too, marked as waiting. So if you file one at the far end of a trail and the history looks unchanged, you can see the walk is still on its way.

You can still fix a recorded walk's photos from there. Expand the walk and each photo has Remove photo, with + Add photo at the end. A stored photo is never quietly replaced. You add a photo or take one off, so the record shows what happened instead of looking as though it had always been that way. This part needs a connection.

On the trail's own page, an Inspections panel lists the most recent walks, newest first. Each shows the rating it came back with, who did it, what the tread was like and the answer that set the rating. Where an inspection created a maintenance task, the row links straight to it.

Trails → Inspections shows the same records across every trail at once. Open it when your question covers more than one trail: what the crew walked last month, what came back Marginal or Poor, who has been doing the walks, and where somebody found standing water. You can narrow it by date range, trail, rating or who walked it. Needs attention collects everything that came back Marginal or Poor, so you don't have to pick both.

Answers read as they were worded at the time. If you reword a question next season, every past record still reads correctly.

Photos stay with the question they were taken on. The photo of the washout sits under Tread, not in a gallery where somebody has to guess.

Share in the header holds both exports. Each one gives you what you are looking at, filtered exactly as it is on screen. PDF report has one row per walk, on your organization's letterhead. Send it to a land manager who has asked what you have on record. CSV spreadsheet has one row per answer instead, so you can sort individual answers.

Asking your own questions

The middle of the form is set up in Settings → Trail inspection questions. Add a question and word it the way your crew talks. Give it between two and six answers, and mark each answer as Good, Marginal or Poor. The worst answer given sets the inspection's rating. Move questions up and down to change the order they are asked in. Use the standard questions puts our set back.

Two questions are always asked and are not in the editor. The page shows them anyway, so you can see the whole form. Is the trail usable right now? comes first because it is the only answer that changes what somebody does in the next ten minutes. If an inspector is interrupted, you still have it. How wet is the tread? is a measurement, not an opinion. The trail-status engine reads it to work out whether recent weather has left your ground rideable, so we set its wording and its levels.

Editing the questions never changes what has already been recorded. Every inspection keeps the questions as they were worded on the day it was done. A question you reword this season still reads correctly against last season's walks. A question you remove stays on the records that answered it. It just stops being asked.

Changes reach the Field App the next time it opens with a connection.

Common gotchas

• The trail's status is yours to set. An inspection recommends and records. It never opens or closes the trail by itself. Where weather automation is on, the automation sets the status, and if two things set it you could not tell which one changed it last.
• A trail nobody has walked is scored on the forecast alone, exactly as before. Nothing changes for an account that never inspects.
• Filing late is fine. Walk a trail with no signal and the inspection waits in the queue. It records the day you were on the trail, not the day it reached us, and the engine uses the day of the walk.
• A back-dated record does not move anything backwards. Catching up on a paper log will not overwrite the date of a more recent walk.