Keeping a Tree Health and Disease Scouting Log

What a scouting log is for

Integrated pest management starts with looking. Penn State's extension overview of orchard IPM puts scouting first in its sequence: the first activity is gathering information about the environment, the pest and the crop, and only then does a decision get made. Michigan State publishes a pocket scouting guide for identifying apple pests and damage in the field, and Utah State's home orchard pest guide tells growers to check their trees at least once a week during the growing season. British Columbia's RDOS pest program goes further and gives residential growers a printable weekly diary for logging tree health and pest sightings by hand.

Most growers who lose this habit lose it for one reason: the log never pays for itself. Entries go in a notebook, nothing comes back out. A scouting log pays when three things are true about it:

• It is per tree, not per block. You can only diagnose by comparing a tree against its own past and against its neighbors. Block-level notes cannot do that.
• It is photographic. A photo of the symptom at first observation, dated, is worth more than any written description, because symptoms change and your memory of "small brown spots" is not evidence.
• It closes the loop. Each entry records what you did and what happened next. That is the part that turns a diary into decision-making skill.

The per-tree record, field by field

For every tree, keep a running list of dated entries. Each entry carries:

• Date and pass (routine weekly walk, post-rain check, pre-harvest pass)
• Photos: at least one wide shot of the whole tree and one close-up of the symptom
• Health grade on a simple scale (healthy, watch, treatment needed, critical)
• Symptom described in one line, with location on the tree (new flush, interior leaves, trunk base, fruit)
• Action taken (spray, prune-out, fertilizer change, nothing, watch another week)
• Outcome noted at the next pass (resolved, same, worse)

That structure is deliberately boring. Its power is cumulative: after two seasons you can answer questions like "does this tree get leaf problems every year after heavy rain?" and "did the copper spray actually shorten that outbreak last time?" Neither question has an answer without the log.

The routine: how often, and what triggers more

1. 1Weekly during the wet season and the fruiting season; fortnightly in dry dormancy. Humid, warm periods are when fungal and bacterial leaf diseases move fastest in tropical orchards.
2. 2Extra pass within 48 hours after any major weather event (storm damage, hail, prolonged rain). Wounds and wet canopy are how problems start.
3. 3Photograph anything new before you touch it. You want the before-evidence, and the close-up needs to be in focus and fill the frame.
4. 4Sort the worklist worst first. The point of grading every tree is triage: treat the criticals this week, watch the watches, leave the healthies alone. Spraying the whole block because three trees are sick is the expensive version of not having a log.
5. 5Note outcomes at the next pass, even when the answer is "same". An intervention with no recorded outcome is an experiment you threw away.
6. 6Keep trap or lure counts in the same log where you run them. Extension guides for tree fruit recommend checking traps weekly and recording the numbers for future evaluation, because a season of counts is what lets you judge whether this year's pressure is normal or worth a reaction.

A worked example: three weeks in a small mango block

Eight 'Keitt' and 'Nam Doc Mai' trees, wet season, weekly walks. The log reads (illustrative):

• Week 1: Tree K4, new flush, close-up photo shows small irregular black spots on young leaves. Grade: watch. Note: first rain in 3 weeks occurred 4 days ago. Action: none, re-check.
• Week 1: Tree K7, interior leaves yellowing and dropping, trunk base photo shows standing water at the collar after irrigation. Grade: treatment needed. Action: move the emitter away from the trunk, re-check.
• Week 2: Tree K4, spots have enlarged and now show a dark margin with a lighter center on several leaves; two neighboring trees (K3, K5) show first spots on new flush. Grade: treatment needed. Action: prune out affected flush on K4, fungicide per label on K4 and the two new cases, photos filed of all three.
• Week 2: Tree K7, yellowing has stopped spreading; no new leaf drop. Grade: watch. Action: none.
• Week 3: K3 and K5 clean, K4 has one new spotted leaf only. K7 flushing normally. Grades downgraded. Outcome column filled for all four entries.

Three weeks of entries bought you several decisions: you caught an outbreak at one tree instead of eight (two fungicide passes on three trees instead of the whole block), you found an irrigation-collar problem before it became trunk rot, and you now have photo evidence of what early stage looks like on your own cultivars, which next season you will recognize a week earlier. That compounding effect is the actual return on the log.

Common failures, and the fixes

• Logging the block, not the tree. Fix by making the tree the unit of entry; one line per tree per pass, even when the line is "clean". Clean entries matter: they establish what normal looks like for that tree.
• No photos. Written symptom notes are unreproducible. A geotagged, dated photo attached to the tree record is the strongest possible baseline, and it lets you compare this week's frame with last month's on the same tree.
• Diagnosing from one photo and acting immediately. Wrong identity means wrong chemistry. When unsure, photograph, grade conservatively, and hold the intervention one pass unless the symptom is obviously destructive (rapid wilting, trunk cankers, fire-blight-like strikes).
• No outcome column. The log's value compounds through closed loops. If you sprayed, you must look again and record what changed.
• Treating the log as compliance paperwork. The RDOS residential diary and the Penn State spray spreadsheets are useful formats, but the payoff is the decisions, not the document. If entries never change what you do, restructure the log until they do.

Minimum viable log: one line per tree per walk (date, grade, one-line symptom, action), plus a photo whenever the grade is watch or worse. That is under two minutes per tree, and it is the difference between an orchard you remember and an orchard you can reason about.

How CanopyTwin fits in

CanopyTwin's Arborist is built for exactly this loop. You photograph a leaf, a bug or a whole tree; the AI grades every tree from healthy to critical, ranks the whole farm worst first as a worklist, and backs each diagnosis with what is visible in the frame and a suggested fix. You agree or overrule in the field, so you stay the arborist of record, and every photo and grade lands on that tree's permanent record next to its survey history. The iOS app is on the App Store, and every plan starts with a 30-day free trial, no credit card up front.