Skip to content

Records

Did the Damage Start Before the Freeze? Keep a Tree Photo Record

A dated, same-spot photo timeline for each tree, with protection and temperature sources, shows whether a problem was already recorded before a cold event. It narrows the question; it does not diagnose.

Updated November 9, 2026 · 8 min read

Sooner or later a tree shows damage after a cold night and you ask the obvious question: did the freeze do this, or was it already there? If you did not photograph the tree or write down what it looked like before the cold, you are left guessing. A before-and-after observation record fixes that. It is a short, dated photo timeline for each tree, taken from the same spots, with a note of what protection the tree had and what temperature readings you can point to. It will not tell you the cause. It will tell you when the change was first recorded, between which two dated visits, and that narrows the list of causes worth chasing.

One grower's version of the question

The question comes up in the long "Great Seattle cold-hardy avocado trial" thread on the Growing Fruit forum. On page 34, the poster swincher (29 January 2024) asked for help with a tree in the greenhouse. The post says the damage "seemed to come on suddenly in the last month" and that the poster was not sure whether it predated the freeze, because that corner of the greenhouse rarely gets a look. The poster wondered whether there had been a cold pocket there. The post includes two photos, one of the tree as it looked then (a graft of Wilma, per the post) and one from a couple of months earlier.

The post also says it did not really look like freeze damage, and hopes it was not a systemic infection, naming Verticillium as a possibility because another tree had been suspected of it and culled the year before. A second poster, Marsjaner, replied that the leaves of their own Lila had similar brown discolouration, first seen in autumn with a lot of light and temperatures around freezing, and that this combination was where it came from, in their view. swincher answered that this case had not had much light, that the coldest greenhouse reading was above freezing, and that the corner, even if colder than the sensor, could not have been much below freezing.

We are not telling you who was right, and the thread page does not settle it. What it shows is how much a grower can reason from two dated photos, one sensor and a note about where the light falls, and how much stays open. The poster thought it appeared within the last month, and the photos show a change since a couple of months earlier. They do not let anyone name the cause.

What photographs can and cannot do

A photo shows appearance on a date. Similar-looking leaves can have different causes, and the two growers above compared leaves that looked alike and still disagreed about the cause. Treat the photo timeline as a way to sort observations by date, not as a diagnosis. Its job is to answer three narrower questions.

  1. Was the change already visible before the cold event? If the earlier photo already shows it, the cold was not the start, though it may still have made things worse.
  2. Did it spread, stall or reverse after the cold event? Compare later photos of the same spot with the pre-event one.
  3. Where on the tree did it appear first? Position (low or high, near the graft or far from it, one side or all round) is something you can only see by comparing like with like.

If the answer matters, for example before you cull a tree, take the dated photos to someone who can examine it in person, such as your local extension office or a plant clinic. Your record gives them a timeline instead of a guess.

The record, field by field

For what to record before calling a tree cold hardy at all (lowest temperature, duration, cultivar and rootstock), see our piece on what to record before calling a citrus tree cold hardy. This record is different: it is a timeline, built to be read after the event.

  • Tree ID: one code on the tag, the map and every photo, never renumbered.
  • Cultivar and rootstock, and what the graft is, written the same way each time. In the thread, the tree is described as a graft of a named variety, and on the same page the poster reports ungrafted seedlings separately from grafted trees, so the distinction matters to the people keeping the records.
  • Graft location: height of the union above the soil, and whether soil or mulch is mounded over it. Photograph the union itself.
  • Photo spots: two or three fixed positions, for example the whole tree from the north, a close shot of the graft, and a close shot of the worst-looking leaf or branch. Mark the standing spot with a stake or write the distance and direction.
  • Photo date and time, and the light (sun, overcast, flash). The thread's mention of diffused light in one corner is a reminder that light changes how leaves look.
  • Protection used on each date: none, covered, wrapped, mounded, heated, or in a greenhouse, and when it went on and off. "None" is a real entry, not a blank.
  • Temperature source: which thermometer or sensor, where it is relative to the tree, and what it read. Say if it is a min/max thermometer and when you last reset it. In the thread, swincher notes the corner might have been colder than where the sensor sits, and that is exactly the caveat to write down.
  • Weather event note: date of the cold night, and the readings from your own sensor and any nearby station you used, with the station name.
  • Recovery notes: what grew back, from where, and when. Date each one, including "no change" and "nothing yet".

A template you can copy

  • Row 1, autumn before any cold: Date | Tree ID | photo spots taken (whole tree, graft close-up, one branch) | protection: none | sensor: not yet cold | what I see: no browning visible.
  • Row 2, day after a cold night: same date, ID and spots | protection on | min reading and the sensor it came from | describe only what you see, no cause.
  • Row 3, two weeks later: same spots | protection | reading, if any | compare with row 1 and say what changed.
  • Row 4, spring new growth: same spots | where growth started, and from which part of the tree.

When to take the photos

The photo that matters most is the one you took before anything happened. Take a baseline set for every tree in autumn, before the first cold forecast, and again at a regular interval through winter, because the thread's example shows how easily a corner goes unvisited. A fixed schedule, such as the first of each month, is better than photographing only when something looks wrong. Then take a set the day after any cold night, and again about two weeks later, since some changes may take time to show. So do not decide on day one, and check with your local extension office for what is usual for your species.

Reading the record after one season

  1. Line up each tree's photos by date and find the first one showing the change.
  2. Mark the cold night on the same line. Is the first visible change before it, just after it, or weeks later?
  3. Check the protection column. If trees with different protection changed differently, you have a lead. If every tree is the same, you have little to compare.
  4. Check the sensor column against the location of the tree. A reading from a spot far from the tree is a hint, not a measurement of the tree's own conditions.
  5. Read the recovery column last. What regrew, and from which part, is often more informative than how the damage looked on day one.
  6. Write one cautious sentence per affected tree, for example "first visible between the autumn and January photos; cold night in between; cause not established".

Notice that the last sentence can honestly say "cause not established". That is a good result. It stops you drawing a conclusion the record cannot support.

Reading it after several seasons

  • Patterns by position: if the same corner or the same side of several trees changes first each year, look at that place (light, airflow, drainage), not at the individual trees.
  • Patterns by graft: if the same cultivar or rootstock repeats the same early change, that suggests an association worth checking against location and treatment, since the three can coincide.
  • Patterns by protection: compare years with and without protection on the same tree.
  • Changes with no cold event: if a tree shows changes that line up with no cold night in any year, the cold may not be the story, and you will want help from someone who can examine it.
  • Recovery speed: note how long each tree takes to put out growth after a cold event, and whether it varies by year.

Keep old photos even for trees you remove, filed under the old Tree ID. The thread above mentions a tree culled the previous year and suspected of an infection; a record like this is what lets you look back at such a tree and compare it with the next one that looks similar.

Paper first, an app if you want it

A folder per tree on your phone and a spreadsheet with the columns above is enough, and so is a notebook with a printed photo taped beside each entry. If you would rather see your trees on a map with the photos and notes attached to each one, CanopyTwin does that on iPhone, with a 30-day free trial. It is a convenience, not a requirement, and the timeline works the same on paper.

Turn your orchard into a digital twin

CanopyTwin does the survey-to-map pipeline for you — QR tree tags, a live satellite map, and AI fruit counts from your photos. Start free, no card required.