Plant Variety Database

An open dataset that joins cultivar-level seed-catalog data with USDA hardiness zones and per-zone monthly planting calendars — 1,972 varieties × 13 zones × 12 months, fully sourced, CC BY 4.0.

The hero rows aren't the 1,972 varieties (USDA PLANTS already has ~98K species). They're the joins:

• 20,728 variety × zone planting-calendar entries (indoor sow / transplant / direct sow / harvest windows)
• 21,880 companion-plant pairings with relationship and reason
• 2,327 outbound citations to extension factsheets, breeder pages, and USDA records (92% of varieties carry at least one)
• 1,036 USDA FoodData Central nutrition records joined to growable varieties

The full interactive version — variety pages, planting calendars per zone, companion-plant prose, troubleshooting guides, and per-county zone maps for all 50 US states + 13 Canadian provinces — lives at plants.windrivergreens.com.

Also available on

Host	Link
GitHub (canonical)	github.com/bripatch/plant-variety-database
Kaggle	kaggle.com/datasets/windrivergreens/plant-variety-database
Hugging Face	huggingface.co/datasets/windrivergreens/plant-variety-database

The GitHub repo is canonical and auto-refreshes monthly. Mirrors are kept in sync but may lag by up to a month.

Why this dataset exists

USDA PLANTS gives you species-level taxonomy but no cultivars and no planting calendars. Johnny's Selected Seeds catalog gives you cultivar-level days-to-maturity but no zone-by-zone schedule and no nutrition. NC State Extension gives you growing prose but no structured cultivar database. Hardiness zone shapefiles give you a map but no variety information.

This dataset is the join. 1,972 hand-cleaned variety rows, each linked to a planting calendar for every USDA zone it grows in, with companion plants, common pests and diseases, and (where applicable) USDA nutrition data per 100g — all in flat CSVs with verifiable source citations.

What's in the data

File	Rows	What it is
data/varieties.csv	1,972	One row per cultivar — name, scientific name, days to harvest, plant size, sun/water/soil needs, USDA zones, pest/disease info
data/categories.csv	30	Plant category index (tomato, herb, rose, succulent, etc.)
data/zones.csv	13	USDA hardiness zones 1–13 with temperature ranges, frost dates, growing-season length
data/planting_calendar.csv	20,728	Variety × zone — when to start indoors, transplant, direct-sow, harvest
data/companion_plants.csv	21,880	Beneficial and harmful plant pairings with reasons
data/nutrition.csv	1,036	Per-100g nutrition (calories, macros, vitamins, minerals) from USDA FoodData Central
data/sources.csv	2,327	Outbound citations per variety — links to extension factsheets, breeder pages, USDA records

All files are UTF-8 CSV with a header row, double-quoted escaping for embedded commas/newlines. Foreign keys: every variety_*.csv row joins back to varieties.csv via variety_id or variety_slug. planting_calendar.csv and nutrition.csv join to zones.csv and FDC respectively.

Field reference for varieties.csv (click to expand)

Column	Type	Notes
id	integer	Primary key
category	string	Lowercase slug — joins to categories.slug (tomato, herb, rose, …)
name	string	Display name including cultivar — e.g. Cherokee Purple Tomato 'Cherokee Purple'
slug	string	URL slug — joins to *.variety_slug across other files
scientific_name	string	Genus + species (+ cultivar epithet where known)
description	string	1-3 sentence overview
days_to_harvest	string	Range or value — e.g. 60-80, 70, or null. Parse with regex when filtering numerically.
days_to_germination	string	Same shape as days_to_harvest
plant_height	string	Free-form — e.g. 4-6 ft, 12 in
plant_spacing	string	Free-form — e.g. 18-24 in
sun_requirement	string	full_sun, partial_sun, partial_shade, full_shade
water_requirement	string	low, medium, high
soil_type	string	Free-form
soil_ph	string	Range — e.g. 6.0-7.0
growing_difficulty	string	beginner, intermediate, advanced
is_container_friendly	boolean	true / false
growing_season	string	cool, warm, year_round
sowing_method	string	direct_sow, transplant, both
color, size, shape, flavor_profile	string	Sensory / appearance fields where applicable
culinary_uses	string	Free-form, semicolon-separated
is_heirloom, is_hybrid	boolean	Cultivar provenance flags
usda_zone_min, usda_zone_max	integer	Hardy zone range, 1-13
disease_resistance	string	Free-form
common_pests, common_diseases	string	Semicolon-separated lists
source_database	string	Comma-separated list — nc_state, johnnys, usda_plants
url	string	Permalink to the live variety page on plants.windrivergreens.com

Use cases

A few research adjacencies this dataset enables:

• Climate-zone migration modeling — pair usda_zone_min/usda_zone_max with future-zone projections (e.g. USDA PHZM) to see which cultivars become viable / unviable in a given county over time.
• Agricultural ML training — clean, labeled cultivar data with consistent feature schema for taxonomy, growability, and nutrition classification tasks.
• Garden-app / smart-home / IoT data — drop-in zone-aware planting calendar without re-licensing per-source data per platform.
• Nutrition × growability joins — nutrition.csv + varieties.csv lets you ask "which high-vitamin-K leafy greens grow in zone 4?" in one query.
• Companion-planting network analysis — companion_plants.csv is a 21,880-edge undirected graph (with relationship labels) suitable for graph-DB ingestion or polyculture optimization.
• Education — pre-cleaned, sourced data for K-12 / undergraduate horticulture and ag-science curricula.

Quick start

import pandas as pd

varieties = pd.read_csv("data/varieties.csv")
calendar  = pd.read_csv("data/planting_calendar.csv")
nutrition = pd.read_csv("data/nutrition.csv")

# Tomatoes that mature in under 80 days and grow in zone 7
tomatoes_z7 = varieties[
    (varieties.category == "tomato")
    & (varieties.usda_zone_min <= 7)
    & (varieties.usda_zone_max >= 7)
    & (varieties.days_to_harvest.str.extract(r"(\d+)")[0].astype(float) < 80)
]
print(tomatoes_z7[["name", "days_to_harvest", "growing_difficulty"]])

# When to start each variety indoors in zone 6
z6 = calendar[calendar.usda_zone == 6]
print(z6[["variety_slug", "indoor_sow_start", "outdoor_transplant_start"]].head(20))

# High-vitamin-K leafy greens that overwinter in zone 4
leafy_z4 = varieties[varieties.category.isin(["lettuce", "kale", "chard", "spinach", "arugula"])]
high_k = nutrition.merge(leafy_z4, on="variety_slug").query("vitamin_k_mcg > 100 and usda_zone_min <= 4")
print(high_k[["name", "vitamin_k_mcg", "usda_zone_min", "usda_zone_max"]])

See examples/queries.py for more.

What's NOT in this export

This dataset is the structured index. Long-form editorial content stays on the site:

• Per-variety growing guides, history, harvest/storage notes
• Troubleshooting trees (with HowTo schema)
• Companion-planting prose narratives
• Succession-planting schedules
• Per-cultivar source citations rendered in context

Every row has a url column — link out for depth.

Sources & verification

Every variety in this dataset is backed by at least one real data source — no AI-generated plant facts. Two layers of provenance:

Per-variety data origin — the raw feeds that built each variety's record (source_database column on varieties.csv; most rows combine 2-3 sources):

Source	Varieties	Contributes
NC State Extension	1,794 (91%)	Zones, height, light, growth rate, growing guides
Johnny's Selected Seeds	939 (48%)	Cultivar-level days to maturity, spacing, disease resistance
USDA PLANTS Database	506 (26%)	Species-level characteristics

Outbound citations — verifiable links to original factsheets/data per variety (sources.csv, 2,327 entries; 1,822 of 1,972 varieties (92%) carry at least one):

Source type	Distinct varieties
USDA FoodData Central (nutrition)	1,036
Breeder pages (Johnny's Selected Seeds)	799
Extension factsheets (NC State, UGA)	481
Botanical gardens (Missouri Botanical Garden)	11

Updates

This dataset auto-refreshes from the live production database monthly via GitHub Actions (1st of each month, 08:00 UTC). The live site at plants.windrivergreens.com updates continuously; this snapshot batches changes into a single monthly commit.

For academic use, pin to a specific tagged release — those are immutable. The main branch tracks the latest export and will change over time.

License

Creative Commons Attribution 4.0 International (CC BY 4.0)

Free to share, adapt, and build on for any purpose — including commercial — as long as you credit Wind River Greens with a link back to plants.windrivergreens.com.

Suggested attribution

Plant variety data from Wind River Greens Plant Database (CC BY 4.0).

Citation (BibTeX)

For academic use, cite a specific tagged release — each release is an immutable snapshot:

@misc{windrivergreens_plantdb_2026,
  author       = {{Wind River Greens}},
  title        = {Plant Variety Database: A cultivar-level dataset with USDA hardiness zones and per-zone planting calendars},
  year         = {2026},
  version      = {1.0.0},
  url          = {https://github.com/bripatch/plant-variety-database},
  howpublished = {GitHub repository},
  note         = {CC BY 4.0. Live tool: \url{https://plants.windrivergreens.com}}
}

Issues, corrections, contributions

Spot a wrong zone, a misclassified variety, or a missing companion-planting relationship? Open an issue — corrections are very welcome and flow back into the live site.

About

Wind River Greens is a small microgreens farm in Milton, Georgia. The plant database started as a tool for our customers and grew into one of the larger free, sourced, zone-aware variety datasets on the open web.