Plant Counting in Quadrats: A Botanist Scenario

How should a quadrat survey be set up before fieldwork?

Maren spends an evening at her desk mapping the protocol onto the app before she ever kneels in the grass. The app gives her categories nested up to three levels, counters inside them, and tags inside each counter.

She settles on:

• Category level 1: the site, "Low Meadow".
• Level 2: the transect, "T1", "T2", "T3".
• Level 3: the visit, "Visit 1 May", "Visit 2 June", "Visit 3 July".
• Counters: one per quadrat, named "T1-Q01" through "T1-Q10".
• Tags: one per indicator species.

She pins the current visit's category so it sits at the top of the list. Counters can be duplicated for the next visit by importing a backup as a copy, but she prefers creating each visit's quadrats fresh, which forces her to check every name against the plot map.

A colleague suggests the opposite layout, with species as counters and quadrats as tags. Maren tries it on paper and rejects it, because her analysis treats quadrats as the sampling unit and each export row should be one quadrat.

Species tags and their settings

Each quadrat counter gets the same eight tags, in the same order, with the same colors: yellow rattle in yellow, knapweed in purple, bird's-foot trefoil in orange, and so on. Matching colors to flowers is a small trick that pays off later, when she is tired and reading the screen at an angle.

Every tag has its own step, value, optional target, timer, stopwatch, location, barcode and notes. For her stem counts the step stays at 1. For one grass she records in clumps of five, she sets that tag's step to 5 and writes "step 5 = one clump" into the counter description.

The tag location is typed text, "Low Meadow T1 Q01", because GPS on a phone is not precise enough to separate quadrats a few metres apart. She does store a GPS fix on the first quadrat of each transect as a rough anchor for anyone relocating the plots later.

A day in the field

On the first visit, Maren works along T1 on her knees, frame in place. She opens the quadrat counter and uses List view, where all eight species sit as rows with their own minus and plus buttons and the running values.

For the most abundant species she switches to Single view. One big circular counter fills the screen and she taps without looking closely, eyes on the stems. The tag card below shows the location and the most recent note, so she can confirm she is in the right quadrat.

Counting with the volume keys

On the second day she enables volume-key counting in settings. It is off by default. She maps volume up to plus and volume down to minus, gives each its own short vibration, and keeps the sound muted. This lets her hold the phone in one hand and part the vegetation with the other. The counting continues even with the screen off, through a foreground service, and the lock-screen notification also carries minus, plus and stop buttons. On her phone the system volume bar still flickers briefly, which she learns to ignore.

Locking finished quadrats

When a quadrat is done she locks its counter. Kneeling, standing and pocketing a phone produces a lot of accidental touches, and a lock stops them from changing finished data. Unlocking takes a single deliberate tap.

Photos of tricky specimens

Some plants cannot be identified with confidence in the field. Maren photographs them on the relevant tag, which holds up to six photos. In the photo editor she draws a quick arrow with the brush, adds the quadrat code as text and rotates the image upright. Later, the all-photos gallery shows every image with its tag label and date, which makes it easy to send a batch of uncertain specimens to a colleague who knows the grasses better.

She attaches notes to individual count events as well: "two plants grazed, counted stems only" or "frame shifted 10 cm, re-placed". These end up in exports alongside the counts.

The evening routine

Back at the car, Maren runs a short check:

1. Open the All Notes screen and read every note from the day, sorted by time.
2. Open history for any quadrat with a surprising total. Each event shows the time, step, before and after values, tag label and icons for notes or photos.
3. Compare a few totals with the paper backup sheet.
4. Make a backup ZIP with photos, choosing only the current visit's category, and copy it to her laptop.

History cannot be searched, so she works quadrat by quadrat. On day one she finds a block of minus taps at the end of T1-Q04 that she does not remember. The timestamps show them landing while she was walking back to the car, which is exactly the problem locking solves, and from then on she locks every quadrat.

Working with a field assistant

On the June visit a student, Theo, counts T3 on his own phone while Maren does T1 and T2. There is no live sync between phones, so the plan is simple. Theo works only in the T3 category, and at the end of the day he creates a backup ZIP of that category.

When Maren imports it, the preview lists the counters and asks how to handle duplicates: replace, copy or skip. Since T3 does not yet exist on her phone, nothing clashes. On a later visit, when both phones contain an older copy of T3, she chooses replace for Theo's fresher counters and keeps the ZIP files as a dated archive.

They also agree to swap phones for two quadrats and count them blind, then compare. That gives Maren a rough idea of observer differences without designing a formal study.

How can quadrat counts be compared between visits?

Between visits, the dashboard on each quadrat counter gives Maren a quick picture: the total, increments against decrements, a seven-day activity chart and a pie chart by species. The pie chart is not analysis, but it shows at a glance that T2-Q07 is dominated by one species while its neighbours are mixed, which prompts her to check the plot boundary on the next visit.

She does the real comparison in a spreadsheet, not on the phone, because visits live in separate categories and the app is not built for cross-visit statistics.

End of season: analysis

In August Maren exports each visit's counters as CSV. In the export options she ticks Summary and Locations, adds Notes for the record, and leaves Events off because she does not need every individual tap. The Summary gives one total per species per quadrat, and the Locations section carries her typed quadrat codes.

She loads the files into R, binds them by visit, and reshapes them into a quadrat-by-species matrix. An Excel export goes to the landowner, and an HTML report with embedded photos goes into the project archive, since it reads well without any software.