Your garden doesn't read a calendar

Every arborist in Minnesota knows the rule: don’t prune oaks from April through July. Fresh wounds in those months draw the sap beetles that carry oak wilt, and a red oak that catches it dies within the year.¹

In 2012, the rule ran late. That March the Twin Cities broke six daily record highs and hit 80°F earlier than ever before.² Minneapolis–St. Paul stayed at 60°F or warmer for twelve straight days in the middle of the month — twice the six-day mark the Minnesota DNR uses as its rule of thumb for when oak wilt risk begins. In Rochester, the DNR dates the start of risk that year to March 18.¹

A crew pruning oaks in the last week of March 2012 was following the rule exactly — and, by the DNR’s own rule of thumb, cutting with the risk window already open.

Nobody was careless. They followed the calendar. The calendar was the problem.

Plants keep time in heat

A plant has no idea what the date is. In spring, what it responds to is accumulated warmth, and horticulturists measure that warmth in growing degree days (GDD). Each day, average the high and the low and subtract a base temperature — usually 50°F for landscape plants and insects — counting zero when the day falls short. The running total from a fixed start date is the number the plant, and everything that feeds on it, is actually counting.

That’s why a lilac doesn’t bloom on a date. The Minnesota DNR’s State Climatology Office has logged Twin Cities lilacs since 1999. In 2012 they first bloomed on April 14, the earliest in the record. In 2013 they bloomed on May 25. Forty-one days apart, in back-to-back springs: a record-warm March pulled one season forward, and a cold, snowy April — the airport didn’t reach 60°F until April 26 — shoved the next one back.³

Two springs at Minneapolis–St. Paul

Heat banked since January 1, in growing degree days (base 50°F)

• 2012
• 2013

02505007501,000

Apr 1420122013

Mar 1Apr 1May 1Jun 1

• 2012 lilacs bloomed Apr 14, at 188 GDD
• 2013 lilacs bloomed May 25, at 280 GDD

View the data

Date	2012	2013
Mar 1	0	0
Mar 15	20	0
Apr 1	162	0
Apr 14	188	0
May 1	243	62
May 15	407	163
May 25	575	280
Jun 1	663	373
Jun 15	953	567

Count heat instead of days and the picture sharpens. We ran the record: across the 27 springs from 1999 to 2025, predicting lilac bloom from accumulated heat missed by about three days on average. Predicting it from the usual calendar date missed by about seven — and by 29 days in 2012.⁴

The same gap runs through everything a crew does in spring. Pine needle scale crawlers hatch at roughly 300 to 450 GDD, the short window when a treatment works.⁵ At MSP that window ran May 5–18 in 2012 and May 28–June 9 in 2013. In a typical year it falls around May 23 to June 4. A spray booked for the usual weeks would have missed 2012 entirely.

The schedule gets the truck right, and the biology right only by accident.

Experienced gardeners know this in their gut. They walk the property, read the plants, and adjust. That instinct is the most valuable thing an estate firm owns. It’s also trapped in a few people’s heads — and it doesn’t scale, doesn’t take a vacation, and walks out the door when they retire.

The schedule still runs on dates

Most of the software landscaping firms run on schedules work by date: prune in week 11, treat in week 21. Those tools grew out of routing and invoicing, and they treat a high-touch estate garden the way they’d treat a mowing route — a stop on a map, on a day, repeated.

For much of the green industry that’s fine. Grass gets cut on Tuesdays. For firms managing dozens of species on a single property, where one mistimed cut can cost a client relationship, it’s quietly expensive: the crew arrives on the right day for the route and the wrong day for the plant.

The knowledge to do better already exists. Extension services publish degree-day thresholds. The USA National Phenology Network forecasts pest timing. The DNR points arborists to a thermal model that lets them “extend their productive work in many years while not exposing oaks to infection risk.”¹ But that knowledge lives in a browser tab and a horticulturist’s memory, not in the schedule the crews actually follow.

What we’re building

Kloro runs an estate horticulture firm’s operations, and it raises an advisory when the biology opens a care window, not when the calendar turns. It pulls daily weather for every property, tracks the heat each one has banked, and checks it against the windows it knows for the plants and pests on it. So the advisory says bronze birch borer flight starts this week — protect the specimen birches, not it’s May, so it’s probably time, and the firm decides when the crew goes.

Degree days alone don’t run a business, so Kloro also carries the unglamorous spine: visits, crews, field reports, client preferences, invoices, payments, and the “can we take on another client?” math owners can never quite answer. The operational depth is why a firm switches. The biology is why it doesn’t switch back.

The goal is easy to say and hard to build: take the judgment in your best horticulturist’s head and make it something the whole firm can run on.

Why it matters now

Growth is exactly where the calendar breaks. One brilliant lead horticulturist can keep 30 properties in their head. They can’t keep 300. The knowledge doesn’t copy, the vacation never comes, and the expertise becomes the ceiling.

There’s a better answer than hiring another genius and hoping. Encode what the plants are doing, put it in front of the crew at the right moment, and let the firm grow on systems instead of heroics.

Estate clients aren’t paying for a crew that showed up on the scheduled day. They’re paying for a garden that’s right. In spring, getting it right means counting what the plants count — and what they count is heat.

Next in Field Notes: How to read growing degree days — what they actually measure, and where they mislead.

Sources

1. Minnesota DNR Forest Health Program, Oak Wilt Guide, April 2026 — the April–July prevention window, red oak mortality, start-of-risk dates for 2009–2025, the six-day 60°F rule of thumb, and the thermal model. ↩ ↩² ↩³

2. Minnesota DNR State Climatology Office, Top five weather events of 2012. The twelve-day run (March 13–24) is from NOAA daily highs at MSP. ↩

3. Minnesota DNR State Climatology Office, Twin Cities phenology record, lilacs blooming, 1999–2025; the 2012–2015 dates come from a variety of lilacs in the area. The April 26 date is from NOAA daily highs at MSP. ↩

4. Kloro analysis of NOAA daily highs and lows at Minneapolis–St. Paul International Airport (ACIS). Growing degree days use base 50°F and the simple-average method from January 1. Each year was predicted from the other 26: the heat forecast used their average GDD at bloom, the calendar forecast their median bloom date. ↩

5. USA National Phenology Network, Pine needle scale forecast — first-generation crawlers at 298–448 GDD, base 50°F, from March 1. The window dates are Kloro’s calculation from the same NOAA record. ↩