Insights and reports from PlantData.Live with ChatGPT, Claude or Copilot
As an arborist, you ask questions in plain language and get answers based on the measurements and notes for the trees you monitor. No more clicking through dashboards or retyping numbers.
PlantData.Live collects live data on the trees and plants you look after as an arborist, from young plantings to valuable trees on a construction site or in the city. That includes sensor measurements, such as trunk diameter, soil moisture and weather, as well as the notes and observations you log during a site visit. Everything comes together on one platform.
1. Algorithms on our platform alert you the moment you need to step in, for example when water stress is building up. That keeps the trees and plants in your care in good condition.
2. As an arborist, you also need to plan and report. That is what this article is about.
You plan watering rounds and site visits, you keep track of what happened, and you report to your client, such as a city, a contractor or an estate manager. This expert article explains how to organize that work more efficiently with AI assistants such as ChatGPT, Claude or Microsoft Copilot. We give these tools secure access to all records in the platform, through a protected connection (a so-called MCP server) that can only read. The AI consults your sensors, dashboards, irrigation schedules and logbooks, but cannot change or control anything.
Interpreting and deciding: that is your job as the expert.
What this article covers
Five tasks from an arborist's week, and ideas for working with an AI assistant.
1. The morning check: how are my trees doing?
Without AI: you log in, open the dashboard for each zone, look at the charts and judge whether anything is off. With several locations, you click through each one to take in all the information. Or you drive by.
There is one reference tree with soil moisture and dendrometer sensors. The data from the past week shows:
- The daily rhythm is healthy. The trunk is thickest around 10 a.m. and thinnest around 6 p.m. Nighttime recovery is good.
- There was a dip on September 16. That night the trunk did not fully recover, but the next day it was back to normal. The water deficit did not build up further.
- There is enough soil moisture: no reason to intervene with this weather forecast.
This link takes you straight to the source data in your portal
What you gain: one answer shows what is normal and what stands out, with the date. You only dig deeper where needed.
2. Assessing a tree's condition
Without AI: you assess condition visually, based on experience. Reading a chart and translating it into "healthy or not" takes knowledge of sensor technology and derived values, knowledge not everyone on the crew has. A photo history of a tree in our platform can help you assess how it is evolving. But that takes time.
I can't compare it with other trees: the holm oak is the only measured tree at this location. What I do see is that it appears to be in good condition:
- It recovers every night, so there is no structural drought stress.
- It transpires actively, as you would expect from a healthy tree.
- It is still growing slightly, about +74 µm in 30 days. That is plausible for an evergreen species in September.
If you really want to compare, install a second dendrometer on a similar holm oak, or let me analyze this tree's full history.
What you gain: a well-founded interpretation in plain language that you can pass on to a colleague or client. The AI is also honest about what it cannot conclude from the data.
3. Matching watering to the tree itself
Without AI: young plantings and valuable trees often get water on a fixed schedule. If you irrigate automatically with our portal, the routine looks at soil moisture and the weather forecast. As an expert, it remains important to fine-tune an irrigation routine. But visual signs of water stress can take up to 2 weeks to appear.
I can't derive a fixed frequency from the data, because that depends on rain and sun. But I can clearly see when the tree was short of water this summer:
- July 1–23: the tree kept growing normally. No extra water was needed.
- July 23 – August 16: growth stalled for more than three weeks, and daily shrinkage rose to 315 µm. This is drought stress: water was needed during this period.
- From August 17 the tree recovered strongly, after rain or watering.
Advice: An alert for this tree makes sense as soon as daily shrinkage exceeds 200 µm. Then water thoroughly, about once a week, as long as it stays dry.
What you gain: you water when the tree asks for it, not when the calendar says so. That saves trips and water and prevents weeks of lost growth. After this analysis, you set up the trigger as an automatic notification in PlantData.Live.
4. Turning the irrigation logbook into a report appendix
Without AI: retyping your crews' watering logs in Excel, adding up totals, weeding out errors and attaching the result to the report or invoice for your client.
The overview is ready as a Word file: Irrigation_Beringen_week28_2026.docx.
- 2 registrations on 6/7 and 9/7: both confirmed executed
- Total recorded volume: 294.6 liters
- The automatic irrigation schedule was off that week
Pay attention: There is a second note on 9/7 with missing quantities! Probably a data entry error. That is also mentioned in the document..
What you gain: a clean table and summary in seconds, in the file format you need. The AI doesn't just add things up blindly; it flags duplicate or incomplete entries before they end up in a report.
5. Updating the biweekly report
Without AI: the same chore every two weeks. You open the previous report, export new charts, retype the numbers, rewrite the conclusion and check that no old date slipped through. For construction sites, dewatering projects or aftercare contracts, that quickly adds up to half a day per report.
I read your report and recalculated the same indicators for September 10 through 23. The structure, layout and standard text are kept.
| Indicator | Aug 27 – Sep 9 | Sep 10 – Sep 23 | Trend |
|---|---|---|---|
| Average daily shrinkage | 94 µm | 117 µm | ↑ +25% |
| Maximum daily shrinkage | 148 µm (Aug 27) | 187 µm (Sep 15) | ↑ |
| Net growth (morning maxima) | +90 µm | ± 0 µm | growth levels off |
| Accumulating water deficit | no | no | = |
New conclusion: "Trunk growth is leveling off, which is normal at the end of September. Daily shrinkage increased, but the tree recovers every night. No action is needed; the next check follows in the report for September 24 – October 7."
Updated: the period in the title and header, four key figures, the chart and its caption, and the conclusion. Left unchanged, for you to review: the photos and your site visit notes.
What you gain: from half a day to a few minutes of proofreading. Standard reports in your template, with your emphasis. You remain responsible: the AI does the routine work and tells you what it changed.
Try it yourself?
As an arborist, you need three things:
- an account on PlantData.Live with your sensors, or simply to log notes;
- an AI assistant that supports MCP connections, such as Claude, ChatGPT or Microsoft Copilot;
- the connection to our MCP server, which we set up together with you.
The connection is read-only. The AI does not control any valves and does not change any project settings. Decisions stay with you.
More from our knowledge base
What is a dendrometer?
How to derive growth, shrinkage and water stress from trunk diameter.
What is vapor pressure deficit (VPD)?
Why the evaporative demand of the air drives water stress.