← All Side Quests
Oct 11, 2026Data

Before You Plant

Compare the sun, water, warmth, and time recorded for 140 edible plants.

Try the project ↓
How to use this project

Help

How to explore

The GROW Observatory assembled broad growing requirements for edible plants used across European climate zones. This project compares the recorded categories; it is not a local planting calendar or food-safety guide.

  1. Begin with one useful group

    Show the 19 plants labeled hardy or very hardy with low or very low water need.

  2. Choose your own conditions

    Change the sunlight, water, cold-tolerance, and harvest controls, then choose a matching plant.

  3. Compare the recorded wait

    Read each harvest bar from its earliest to latest recorded day.

  4. Connect the labels

    Choose a water-by-cold-tolerance cell, then see where those plants fall by soil pH and harvest time.

Days to harvest
The source's estimated growing time before harvest; this project draws only complete minimum-to-maximum ranges.
Cold tolerance
The source's very hardy-to-very tender class, not a USDA hardiness-zone rating.
Soil pH
A 0–14 measure of acidity and alkalinity; lower numbers are more acidic, and 7 is neutral.
Full sun
The source's highest-light category; the extract does not give one universal number of hours.
Not recorded
The field is blank or unavailable in this extract; it does not mean zero.
Side Quest / DataStart with 19 plants labeled hardy or very hardy and low or very low water, change four plain-language filters, choose any matching plant, compare complete harvest ranges, select a water-by-cold-tolerance cell, and read the linked soil-pH view.
Edible plant records · Europe · 2020

Before
you plant

Compare the sun, water, warmth, and time recorded for 140 edible plants.

The GROW Observatory brought broad growing requirements into one dataset. Use it to compare the records, not as a local planting calendar or food-safety guide.

01 · Choose conditions

Start with what a plant needs.

Each control removes plants whose recorded category does not match. Choose a plant to carry it into every view below.

140 of 140 plants match

02 · Compare time

Some waits are recorded as ranges.

Each bar runs from the earliest to latest recorded harvest day. The filters above change which complete ranges appear; plants without a closed range stay out of this chart.

69matching plants with a complete range
19of all plants end within 75 days
69of 140 have complete harvest ranges
03 · See the pattern

Water and cold tolerance form the largest groups.

Each cell counts all plants with that pair of source labels. Choose a cell to use both labels as filters above.

Cold tolerance ↓
Water need →
Very lowLowMediumHighVery high
Very hardy00
Hardy0
Half hardy000
Tender0
Very tender0000
04 · Read the ranges

Soil pH and harvest time answer different questions.

Move left to right for a higher pH midpoint. Move up for a longer recorded wait to harvest. The filters above change the points; choosing one changes the plant card and range highlight.

180 days900pH 4.56.58.5
Full sun onlyFull sun or shadePartial shade
Keep the comparison in bounds

A short list is a place to ask better local questions.

The source was built for European climate zones, while this extract holds broad plant requirements rather than a location-specific calendar. Soil, weather, variety, planting method, and local guidance can change what works.

Definitions and missing data
Days to harvest
The source's estimated growing time before harvest. This project compares only closed ranges when it draws a bar or point.
Cold tolerance
The source's very hardy-to-very tender class. It is not a USDA hardiness zone.
Soil pH
A 0–14 measure of acidity and alkalinity. Lower numbers are more acidic; 7 is neutral.
Full sun
The source's highest-light category. The extract does not state a universal number of hours.
Not recorded
The selected field is blank or marked unavailable in the extract. It does not mean zero.

I selected the University of Dundee's Edible Plant Database through the allowlisted TidyTuesday catalog, then opened the underlying university record, dataset files, DOI, and CC BY 4.0 license on October 11, 2026.

Why this exists

The question

Which combinations of sunlight, water need, cold tolerance, soil pH, and harvest time appear in the University of Dundee's Edible Plant Database extract, and how does one plant compare with the rest?

What to try

Start with 19 plants labeled hardy or very hardy and low or very low water, change four plain-language filters, choose any matching plant, compare complete harvest ranges, select a water-by-cold-tolerance cell, and read the linked soil-pH view.

Method and limits

About the data

I discovered the source through TidyTuesday's February 3, 2026 Edible Plants entry and pinned the curated CSV to commit `d6b21e5cba464d9970599000b57c8b0823e8b6ca`. The underlying University of Dundee record names Mel Woods, Andy Cobley, Susan R. Verrall, Roy Neilson, Naomi van der Velden, Gerid Hager, and the GROW Consortium as creators. It identifies the University of Dundee as publisher, DOI `10.15132/10000157`, May 2020 as the release date, and CC BY 4.0 as the license for the database files. The source page describes 146 species in the full database. The pinned curated extract contains 140 rows, so every count here uses 140 rather than repeating the broader source total.

I retrieved the 64,814-byte CSV on October 11, 2026. Its SHA-256 is `20c6bb4c2dc9cacf3bf5908d2b9cde753b1db8b7304f1af071c9b06ab2b95b67`. A standard-library Python reducer verifies the byte count, hash, schema, and 140-row count. It normalizes only case, spacing, one `Brassicas` label, one `Very hard` label, and an explicit high-potassium nutrient label into their matching display categories. It parses numeric ranges without filling missing endpoints. Repeated reductions produced the same 28,783-byte snapshot with SHA-256 `5f207d639169c3f838782a26082c951830263567cee613cff310f9d6cc629aee`. The page imports that snapshot and makes no runtime request.

The harvest chart and soil-pH view include only the 69 rows with a complete minimum-to-maximum harvest range. Twelve open-ended harvest values, two single values, and 57 blank or not-applicable values remain visible as text where relevant but are not turned into closed ranges. The soil-pH point uses the midpoint only for position while the selected plant card keeps the full recorded pH range. Counts describe rows in this extract, not the prevalence of plants in gardens or the chance of success.

The original database supported location-aware planting calendars across 12 European climate zones. The curated extract contains broad requirements, not those calendars. Variety, local weather, soil, planting method, pests, and source revisions can change practical guidance. The source also contains occasional broad or inconsistent categories. This project keeps those limits visible, does not rank plants with a hidden score, and does not offer identification, edibility, or food-safety advice.

Skills used

  • Data analysis
  • Data engineering
  • Statistical reasoning
  • Visualization
  • Information design
  • Interaction design
  • Visual design
  • Creative coding
  • Product thinking
  • Editorial storytelling
  • Accessibility
  • Performance
  • Systems design

What the data shows

  1. The pinned extract contains 140 plant rows. Eighty-seven are labeled full sun only, 44 allow full sun or partial shade, six are partial shade, and three allow full sun, partial shade, or full shade.
  2. The source labels 93 of 140 plants medium-water. Nineteen rows pair low or very low water need with hardy or very hardy cold tolerance. These are broad source classes rather than measured local water amounts.
  3. Sixty-nine plants have a complete harvest range that can be compared without filling a missing endpoint. Nineteen of those ranges end within 75 days; cress is the shortest complete range at 7–11 days in this extract.
  4. A soil-pH midpoint can place a plant on one shared view, but the preferred range matters. The interface keeps the selected plant's full pH range beside the point and avoids reducing the records to a single recommendation score.