phenofect 0.1.4

Package for Forecasting and Exploration of Plant Phenology under Climate Change, which has Parameter Examination, Visualization, Clustering and so on... with Phenological & Meteorological Data.

PhenoFECT: Phenology Forecaster and Explorer under Climate Change

What is PhenoFECT?

PhenoFECT is Python package designed to guide the overall analysis procedure for Budding & Flowering Prediction specially tailored to user-target plant. It offers useful functions including, Key Parameter Examination of temperature-based model, Clustering, Visualization, Downloading & Merging Phenological and Meteorological Data... and so on. The Chill-Day model refined by this package demonstrates the higest prediction accuracy for Korean local areas among previously published models. Additionally, PhenoFECT has a broder objective: to become a generalized, open source tool for accurate prediction of plant phenology and to provide scientific insights on phenology shift in many regions affected by climate change.

What is Chill-Day model and how to apply?

• Chilling and forcing model to predict bud-burst of crop and forest species
• Prediction of Blooming Dates of Spring Flowers by Using Digital Temperature Forecasts and Phenology Models

Table of Contents

• PhenoFECT: Phenology Forecaster and Explorer under Climate Change
  • What is PhenoFECT?
  • Table of Contents
  • How to use PhenoFECT?
  • Main Features
  • Description for Embedded Dataset
  • Physiological Background for Plant Phenology
  • Where to get it
  • Useful Readings & Links
  • Contributing to PhenoFECT

How to use PhenoFECT?

Here is the detailed user guide of PhenoFECT.

Main Features

PhenoFECT is designed to specialize in these areas.

• Contains sufficient Embedded Data extracted from Public Data Portal in Korea.
• Easily download and merge various types of phenological and meteorological data into the embedded data set or create your own. Filter and Preprocess data to make it compatible with the package.
• Predict Bud-burst and Predict Flowering simultaneously for multiple regions with Chill-Day Model and Dataset. Highest accuracy for Korean local areas among previously published models.
• Simple application of Hierarchical Clustering based on Chill-Day Model Temperature Time and 2D & 3D t-SNE method for future analysis.
• Select best key parameter sets with Error Heatmap and Error Contourmap visualization based on Mean Absolute Error(MAE) & Root Mean Squared Error(RMSE).
• After select the best fit parameter set, line_graph & simple regression shows how you select parameters well.
• Detailed shape of Chill-Day Model graph for each location & year and Merged Chill-Day Model graph for each Cluster.
• Contains information about the years of occurrence of El Niño and La Niña in Korea, gives plot how the prediction error shifts under climate change.

Description for Embedded Dataset

Data	Division	Description	Period	Reference
daily_meteorological_data	Daily	95 locations & 39 variables	1907-2025 (Maximum)	Public Data Portal in Korea
monthly_meteorological_data	Monthly	95 locations & 31 variables	1907-2025 (Maximum)	KMA
daylen_temperature_data	Daily	95 locations & 4 variables	1907-2025 (Maximum)	Public Data Portal in Korea
OBS_phenology_data	Animal, Plant, Meteorological Phenomena	Main Target (Budding date/Flowering date/Full Bloom date)	1973-2025 (Observed Once A Year)	KMA
cherry_phenology_data	Cherry (budding/flowering/full bloom)	Extracted from OBS_phenology_data	1973-2025 (Observed Once A Year)	KMA
forsythia_phenology_data	Forsythia (budding/flowering/full bloom)	Extracted from OBS_phenology_data	1973-2025 (Observed Once A Year)	KMA
azalea_phenology_data	Azalea (budding/flowering/full bloom)	Extracted from OBS_phenology_data	1973-2025 (Observed Once A Year)	KMA

All embedded dataset can be downloaded from this repository or here. The daily_meteorological_data is not in repository's Embedded_Dataset Folder due to its capacity.

Physiological Background for Plant Phenology

After summer, if the nutrition & weather conditions are satisfied, woody plants prepare next year flowering by differentiation to flower buds. But to prevent flower bud differentiate to flowers in cold winter condition because of transient warm temperature, flower buds come into dormancy state and their flowering control genes maintain bud statement until they get enough cold requirment.

In the Phenology Model, we call the cold requirement as 'Chill-requirement(Cr)'. If the woody plant get enough cold, dormancy releases. From this time, plant needs Heat to differentiate into flowers. After the heat accumulated same amount to Cr, the Budding event happens. We call that as Bud burst. Last, the amount of heat accumulation flower bud differentiate into flower, flowering, is called as 'Heat-requirement(Hr)'.

• Dormancy initiation: The Day when minimum temperature reaches to 5-7℃. (Depends on species)
• Dormancy release: The first Day when Chill accumulation is lower than Chill-requirement.
• Bud burst: Observed Day when 20% of total flower buds in Woody plant get into bud burst.
• Flowering: Observed Day when 3 flowers are observed in a branch.
• Detailed definition and observation rules are guidelines of KMA(Korea Meteorological Administration).

Where to get it

The source code is currently hosted on GitHub at: https://github.com/CSBL-urap/2024-summer-swkim

Installers for the latest released version are available at the Python Package Index (PyPI)

# PyPI

pip install phenofect

Useful Readings & Links

• KMA(Korea Meteorological Administration
• Public Data Portal in Korea
• Chilling and forcing model to predict bud-burst of crop and forest species
• Predicting Cherry Flowering Date Using a Plant Phonology Model

Contributing to PhenoFECT

All questions, bug reports, bug fixes, enhancements, requests, and ideas are welcome.

Feel free to send an email.

• [e-mail]